Pupillometry as a Biomarker of Attention and Memory: Mechanisms, Methods, and Clinical Applications

Abstract

This article synthesizes current research on pupillometry as an objective, non-invasive measure of cognitive processes underlying memory formation. It explores the neurobiological foundations linking pupil dynamics to attention and arousal, detailing how baseline and task-evoked pupil responses predict long-term memory performance. Methodological best practices for experimental design, data preprocessing, and analysis in cognitive pupillometry are reviewed. The article also addresses common methodological challenges and optimization strategies, and evaluates the validation of pupillary measures against behavioral outcomes and their emerging potential in clinical and pharmaceutical development contexts. Aimed at researchers, scientists, and drug development professionals, this review highlights pupillometry's utility in quantifying cognitive load and memory encoding efficiency.

The Neurocognitive Link: How Pupil Dynamics Reveal Attentional Mechanisms in Memory Encoding

Pupil-Linked Arousal and the Locus Coeruleus-Norepinephrine (LC-NE) System

Application Notes

Pupil-linked arousal serves as a non-invasive window into central brain state dynamics, profoundly influencing attention and memory processes. Research consistently demonstrates that pupil diameter is coupled to the activity of the Locus Coeruleus-Norepinephrine (LC-NE) system, a key neuromodulatory circuit regulating cortical state and cognitive function [1] [2]. Fluctuations in pupil size track not only general arousal but also specific cognitive computations such as belief updating in dynamic environments [3] and the precision of sensory representations [4]. Furthermore, moment-to-moment variations in pupil size preceding stimulus presentation can predict subsequent long-term memory performance, underscoring a state of "readiness to remember" [5]. The temporal structure of memory itself is shaped by pupil-linked arousal, with event boundaries triggering phasic dilations that predict the segmentation of experiences into discrete episodic memories [6]. Integrating pupillometry with electrophysiology and pharmacological interventions reveals that different adrenergic receptor subtypes (α-1, α-2, and β) uniquely mediate the coupling between pupil dynamics and cortical state [1] [2], offering novel targets for cognitive enhancement.

Table 1: Key Quantitative Findings on Pupil-Linked Arousal and Cognition

Cognitive Domain	Pupil Metric	Key Finding	Effect Size / Statistics	Citation
Long-Term Memory	Baseline pupil size (pre-stimulus)	Larger baseline pupil size predicts successful later recall vs. forgetting.	Exp 1: Cohen's d = 0.60; BF₁₀ = 72.18Exp 2: Cohen's d = 0.70; BF₁₀ = 49.96	[5]
Temporal Memory	Pupil dilation at event boundaries	Boundary-spanning item pairs are remembered as farther apart in time than same-context pairs.	t(33) = 2.44, p = 0.02, d = 0.42	[6]
Temporal Memory	Pupil dilation at event boundaries	Impaired temporal order memory for boundary-spanning pairs.	t(33) = -4.77, p < 0.001, d = 0.82	[6]
Sensory Precision	Pupil size & dilation rate (slope)	Larger pupil and stronger dilation linked to more precise cortical stimulus representations (fMRI decoding) and higher confidence.	Correlation with decoded uncertainty and confidence (p < 0.05)	[4]
Perceptual Decision Making	Baseline pupil diameter	Periods of increased pupil size correlated with greater trial-to-trial variability in the rate of evidence accumulation.	Computational model parameter linkage (p < 0.05)	[7]
Predictive Inference	Pupil change	Phasic pupil dilation reflects the probability of a change point in a predictive inference task.	Linear regression, p < 0.05 for 27/30 subjects	[3]

Table 2: Effects of Pharmacological Manipulation on Pupil-Cortical Coupling (Rodent Studies)

Pharmacological Agent	Target Receptor	Effect on Pupil Dynamics	Effect on Cortical State (EEG)	[1] [2]
Propranolol	β-adrenergic antagonist	Alters spontaneous pupil dynamics.	Uniquely modulates EEG power spectrum; LC stimulation does not compensate for changes.
Phentolamine	α-adrenergic antagonist	Alters spontaneous pupil dynamics.	Uniquely modulates EEG power spectrum; LC stimulation does not compensate for changes.
Clonidine	α-2 adrenergic agonist	Alters spontaneous pupil dynamics.	Uniquely modulates EEG power spectrum; LC stimulation does not compensate for changes.

Experimental Protocols

Protocol: Probing the "Readiness to Remember" with Baseline Pupillometry

This protocol uses pre-stimulus baseline pupil size to index the preparatory brain state that influences long-term memory formation [5].

• Participants: Recruit adult participants with normal or corrected-to-normal vision. A sample size of approximately 30 is recommended based on previous studies.
• Stimuli: Prepare a set of items for encoding (e.g., 30 words or images).
• Apparatus: Use an eye-tracker with a high sampling rate (e.g., 1000 Hz) to record pupil diameter. Participants should be seated 75 cm from the monitor.
• Procedure:
  • Encoding Phase: Present each item on the screen for a fixed duration (e.g., 2-3 seconds). Maintain a fixed inter-trial interval (e.g., 5-6 seconds) to allow the pupil to return to baseline.
  • Retention Interval: Implement a delay of several minutes between encoding and retrieval.
  • Retrieval Phase: Administer a memory test. This can be a free recall test, where participants write down all the words they remember, or a recognition test, where they distinguish "old" items from "new" lures.
• Data Analysis:
  • Pupil Data: Segment the pupil data for each trial. Calculate the mean pupil diameter during a 500 ms window immediately before the stimulus onset. This is the baseline pupil size.
  • Behavioral Data: For each trial, code whether the item was later successfully recalled/recognized ("Hit") or forgotten ("Miss").
  • Statistical Analysis: Compare the baseline pupil size between "Hit" and "Miss" trials using paired t-tests or Bayesian equivalents. The hypothesis is that baseline pupil size will be significantly larger for remembered items.

Protocol: Assessing Event Segmentation and Memory via Pupil-Linked Arousal

This protocol examines how event boundaries elicit pupil dilations that correlate with the segmentation of continuous experience into discrete memory episodes [6].

• Participants: Recruit adult participants.
• Stimuli: Create a sequence of visual objects (e.g., 32 everyday objects). An auditory context cue (e.g., a tone presented to the left or right ear) precedes each object.
• Event Structure Manipulation: Keep the auditory cue the same for 8 sequential objects to create a stable context "event." Then, switch the cue to the opposite ear to create an event boundary. Repeat this to parse the full sequence into 4 distinct events.
• Procedure:
  • Encoding Phase: For each trial, present the tone followed by the object. Participants make a content judgment on each object (e.g., "indoor" vs. "outdoor"). Continuously record pupil diameter throughout the sequence.
  • Memory Test Phase:
    • Temporal Order: Present pairs of items from the sequence and ask participants to judge which one appeared later.
    • Temporal Distance: Present pairs of items and ask participants to estimate their relative distance in the original sequence.
    • Source Memory: Test memory for the specific auditory context (left/right ear) associated with each item.
• Data Analysis:
  • Pupil Data: Time-lock pupil data to the onset of event boundaries. Use techniques like Principal Component Analysis (PCA) to decompose the pupil dilation response into its temporal components.
  • Behavioral Data: Compare temporal order accuracy and distance estimates for item pairs that span an event boundary versus pairs from the same event.
  • Correlation Analysis: Correlate specific features of the boundary-evoked pupil response (e.g., peak amplitude, latency) with the magnitude of the memory segmentation effects (e.g., the degree of time expansion for boundary-spanning pairs).

Protocol: Dissecting LC-NE Contribution with Optogenetics and Pharmacology

This detailed protocol, adapted from rodent studies, combines optogenetics, pupillometry, EEG, and pharmacology to causally link the LC-NE system to pupil-cortical coupling [1] [2].

• Subjects: Adult Dbh-Cre mice for targeted LC manipulation.
• Surgical Procedures:
  • Viral Injection: Stereotactically inject an AAV vector encoding the light-sensitive channelrhodopsin (ChR2) under a Cre-dependent promoter (e.g., pAAV-EF1a-double-floxed-hChR2-(H134R)-mCherry-WPRE-HGHpA) into the LC coordinates.
  • Optic Implant: Implant an optical fiber above the LC for light delivery.
  • EEG Implant: Implant skull electrodes over the cortex to record EEG.
• Experimental Setup: Place awake, head-fixed mice in a controlled lighting environment. Simultaneously record pupil size (via a camera) and cortical EEG.
• Experimental Design:
  • Spontaneous Arousal: Record periods of spontaneous, phasic pupil dilations.
  • LC-Evoked Arousal: Deliver brief pulses of blue light via the optic fiber to optogenetically stimulate LC-NE neurons and trigger phasic pupil dilations.
  • Pharmacological Manipulation: Systemically administer adrenergic receptor drugs in separate sessions:
    • Propranolol (β-adrenergic antagonist)
    • Phentolamine (α-adrenergic antagonist)
    • Clonidine (α-2 adrenergic agonist)
    • Repeat spontaneous and LC-evoked arousal measurements under each drug condition.
• Data Analysis:
  • Pupil and EEG Analysis: Compare EEG power spectra (delta, theta, alpha, beta, gamma) during spontaneous vs. LC-evoked pupil dilations.
  • Machine Learning: Train a convolutional neural network (CNN) classifier to distinguish between spontaneous and LC-evoked arousal states based on EEG power bands.
  • Drug Effects: Analyze how each pharmacological agent alters the baseline pupil dynamics, EEG power, and the coupling between LC stimulation and its effects on pupil and cortex.

Signaling Pathways and Experimental Workflows

LC-NE Arousal Pathway

LC-NE Experiment Workflow

The Scientist's Toolkit

Table 3: Essential Research Reagents and Materials

Item Name	Type/Category	Primary Function in Research	Example Use Case
pAAV-EF1a-double-floxed-hChR2-(H134R)-mCherry	Viral Vector	Enables optogenetic activation of specific neuron populations; expresses light-sensitive channelrhodopsin in Cre-expressing cells.	Causally probing LC-NE function by injecting into the LC of Dbh-Cre mice [2].
Propranolol	Small Molecule / β-adrenergic antagonist	Blocks β-adrenergic receptors; used to dissect the role of this receptor subtype in pupil and cortical dynamics.	Pharmacologically isolating the β-receptor's contribution to pupil-cortical coupling [1] [2].
Phentolamine	Small Molecule / α-adrenergic antagonist	Blocks α-adrenergic receptors; used to dissect the role of α-receptors in mediating NE effects.	Investigating the role of α-receptors in arousal-related cortical state changes [1] [2].
Clonidine	Small Molecule / α-2 adrenergic agonist	Activates α-2 adrenergic receptors, which often function as autoreceptors to inhibit NE release.	Probing negative feedback mechanisms within the LC-NE system and its effects on arousal [1] [2].
EyeLink 1000	Apparatus / Eye-tracker	High-speed, high-precision measurement of pupil diameter and other eye movements.	Recording task-evoked and baseline pupil size in human cognitive studies [4] [5].
Convolutional Neural Network (CNN)	Analytical Tool / Machine Learning	Classifies brain states or types of arousal based on complex physiological data patterns.	Distinguishing spontaneous from LC-evoked arousal based on EEG power bands [1] [2].
Principal Component Analysis (PCA)	Analytical Tool / Statistics	Decomposes pupil dilation waveforms into temporally distinct components linked to different cognitive processes.	Isolating specific temporal features of pupil responses to event boundaries in memory tasks [6].

The Readiness to Remember (R2R) framework proposes that an individual's physiological state of attentional preparedness before encountering information significantly influences their ability to encode it into long-term memory [5]. This preparatory state, mediated by moment-to-moment fluctuations in attention, directly affects encoding strength and subsequent recollection. Pupillometry, the measurement of pupil diameter, serves as a non-invasive proxy for quantifying this readiness. Pupil size is intricately linked to autonomic nervous system activity, with dilation reflecting the release of norepinephrine (NE) from the Locus Coeruleus (LC) and associated arousal levels [5]. The Adaptive Gain Theory posits that baseline (tonic) pupil size correlates with the firing rate of LC neurons and performance on attentional tasks [5]. The R2R framework leverages this relationship, suggesting that baseline pupil diameter can predict memory success even before a stimulus is presented.

Key Quantitative Findings

The following tables summarize core quantitative findings from recent research investigating the relationship between pupil size and memory performance.

Table 1: Pupil Size and Subsequent Memory Performance (Free Recall)

Experimental Phase	Comparison	Key Statistical Result	Interpretation
Study Phase (during encoding)	Recalled vs. Forgotten Words	BF₁₀ = 86.58, Cohen’s d = 0.61 [5]	Robust evidence for larger pupil size during encoding of later-recalled words.
Baseline (500 ms pre-stimulus)	Recalled vs. Forgotten Words	BF₁₀ = 72.18, Cohen’s d = 0.60 [5]	Robust evidence for "readiness to remember"; larger pre-stimulus pupil predicts recall.
Overall Performance	Mean Accuracy Rate	36.82% (SD = 14.12%) [5]	--

Table 2: Pupil Size and Subsequent Memory Performance (Recognition)

Experimental Phase	Comparison	Key Statistical Result	Interpretation
Study Phase (during encoding)	"Hit" vs. "Miss" Trials	BF₁₀ = 1583.24, Cohen’s d = 0.95 [5]	Very robust evidence for larger pupil size during encoding of later-recognized words.
Baseline (500 ms pre-stimulus)	"Hit" vs. "Miss" Trials	BF₁₀ = 49.96, Cohen’s d = 0.70 [5]	Strong evidence for the R2R effect in a recognition memory test.
Overall Performance	Mean Hit Rate	65.66% (SD = 12.47%) [5]	--

Table 3: The Arousal-Performance Relationship and Pupillary Dynamics

Concept	Relationship with Pupil Size & Performance	Key Reference
Inverted U-Curve (Arousal/Performance)	Performance peaks at intermediate tonic LC activity (arousal) and declines at low and high levels [5].	Adaptive Gain Theory [5]
U-Shaped Curve (Pupil/Memory)	Pupil dilation correlates with enhanced memory up to an optimal threshold, beyond which further dilation impairs recall [5].	Micula et al. [5]
Memory Retrieval	Active free recall of words (without visual stimuli) is associated with significant pupil dilation [8].	Scientific Reports (2018) [8]

Detailed Experimental Protocols

Protocol: Free Recall and Recognition Memory Paradigm

This protocol details the methodology for establishing the R2R effect, as validated in two experiments [5].

• Objective: To determine if baseline and encoding-phase pupil diameter predict subsequent free recall and recognition memory performance.
• Participants: Ninety-five psychology students (or a comparable cohort).

• Apparatus:

  • Eye Tracker: A high-resolution (e.g., ≥ 60Hz) eye-tracking system with infrared illumination for pupil capture.
  • Stimulus Presentation Software: Software capable of displaying text stimuli and synchronizing with eye-tracking output (e.g., E-Prime, PsychToolbox).
  • Environment: A quiet, dimly lit room to control for ambient light fluctuations.

• Procedure:

  • Participant Setup: Calibrate the eye tracker for each participant.
  • Experimental Trial Structure: Each trial follows a fixed sequence with precise timings:
    • Fixation Cross: Displayed for 1,500 ms. The final 500 ms serves as the critical baseline period for pupil measurement.
    • Word Presentation: A single word is displayed for a standardized duration (e.g., 2,000 ms). Pupil size is recorded throughout.
    • Inter-Stimulus Interval (ISI): A blank screen is presented for a variable duration (e.g., 1,000 ms) to reset arousal.
  • Experiment 1 (Free Recall):
    • Participants study a list of 30 words.
    • After a brief distractor task (e.g., solving arithmetic equations for 2 minutes [8]), participants are given 2 minutes to verbally recall as many words as possible in any order.
  • Experiment 2 (Recognition):
    • Participants study a list of 30 "old" words.
    • In a subsequent test phase, the 30 "old" words are randomly mixed with 30 "new" words. For each word, participants indicate whether it is "old" or "new."

• Data Analysis:

  • Pupil Data Preprocessing: Extract pupil diameter for each trial. Apply blink correction and filtering. For each trial, average pupil size during the 500 ms pre-stimulus baseline and during the entire word presentation period.
  • Statistical Comparison: Use paired-sample analyses (e.g., Bayesian t-tests) to compare mean pupil size for subsequently recalled ("Hit") versus forgotten ("Miss") trials, separately for the baseline and encoding periods.
  • Item-by-Item Control: Perform a repeated-measures ANOVA on baseline-corrected pupil size across all words to confirm the effect is not driven by specific word properties [5].

Protocol: Isolating Attention and Working Memory Load

This protocol is adapted from research dissociating spatial attention from working memory (WM) storage [9].

• Objective: To independently manipulate and measure the effects of spatial attention and WM load on pupil size.
• Design: A change detection task with a 2 (Attended Locations: 1 vs. 2) x 2 (WM Load: 1 vs. 2 objects) within-subjects design.
• Procedure:
  • Encoding Phase: An array of objects is briefly displayed. A spatial cue (e.g., highlighting) indicates which location(s) to attend to. The number of objects to be memorized is instructed separately.
  • Retention Phase: A blank screen is shown for a short interval (e.g., 1,500 ms). Pupil size is measured during this phase.
  • Test Phase: A test array is presented, and participants decide if the object(s) in the memorized location(s) has changed.
• Key Outcome: The design allows for the independent quantification of how the number of attended locations and the number of memorized objects affect pupil diameter [9].

Signaling Pathways and Neurophysiological Workflows

Diagram 1: Neurophysiological Pathway from Cognition to Pupil Dilation

Diagram 2: Experimental Workflow for R2R Validation

The Scientist's Toolkit: Research Reagent Solutions

Table 4: Essential Materials and Tools for Pupillometry-based Memory Research

Item	Function/Description	Example & Key Considerations
High-Speed Eye Tracker	Precisely measures pupil diameter at a high sampling rate (≥ 60 Hz).	EyeLink, Tobii Pro; Critical for capturing rapid, task-evoked pupillary responses.
Stimulus Presentation Software	Prescribes experimental paradigm; displays stimuli and records timing.	E-Prime, PsychoPy, PsychToolbox; Must allow synchronization with eye tracker.
Pupillometry Data Analysis Suite	Processes raw pupil data; handles blink removal, filtering, and segmentation.	Custom scripts (Python, R) or software (EyeLink Data Viewer); Requires baseline correction and trial-averaging capabilities.
Controlled Testing Environment	A dedicated space with stable, dim lighting.	A testing booth or room; Essential for preventing confounding light-induced pupil constriction/dilation.
Validated Memory Word Lists	Standardized linguistic stimuli for encoding.	Lists matched for frequency, concreteness, and length; Minimizes item-specific effects on pupil response [5].
Statistical Analysis Software	Performs inferential and Bayesian statistics on pupil and behavioral data.	JASP, R, SPSS; Bayesian methods are particularly useful for quantifying evidence for the alternative or null hypothesis [5].

Task-evoked pupillometry has emerged as a powerful, non-invasive tool for investigating the cognitive processes underlying memory encoding. The pupil's response to cognitive tasks is linked to activity in the autonomic nervous system and serves as an indirect marker of arousal, attention, and mental effort allocation—processes fundamental to memory formation. This application note synthesizes recent evidence connecting pupillary measures to memory performance across different testing paradigms, providing researchers with standardized protocols and analytical frameworks for implementing pupillometry in memory research.

The physiological basis for cognitive pupillometry lies in the brain's locus coeruleus-norepinephrine (LC-NE) system, which regulates pupil size in response to cognitive demands. During high-level arousal, the pupils dilate due to intense secretion of norepinephrine from the LC via sympathetic pathways [10]. Dopamine, released in arousing states such as motivation, reward, or novelty, exerts an excitatory action on the wakefulness-promoting neurons of the LC, resulting in increased alertness [10]. This neurophysiological link makes pupillometry a valuable method for investigating the moment-to-moment fluctuations in cognitive state that predict successful memory encoding and retrieval.

Key Findings and Quantitative Evidence

Recent research demonstrates that pupil diameter during encoding phases robustly predicts subsequent memory performance. The following table summarizes key quantitative findings from recent studies investigating the relationship between pupillary measures and memory outcomes.

Table 1: Summary of Key Quantitative Findings on Pupillometry and Memory Performance

Study Paradigm	Memory Measure	Pupillometric Predictors	Effect Size	Statistical Evidence
Long-term Memory Test (Free Recall) [10]	Recall Accuracy (Mean: 36.82%, SD: 14.12%)	• Larger pupil size during encoding for recalled words• Larger baseline pupil size (500ms pre-stimulus) for recalled words	Cohen's d = 0.61 (encoding)Cohen's d = 0.60 (baseline)	BF₁₀ = 86.58 (encoding)BF₁₀ = 72.18 (baseline)
Long-term Memory Test (Recognition) [10]	Hit Rate (Mean: 65.66%, SD: 12.47%)	• Larger pupil size during encoding for "hit" trials• Larger baseline pupil size (500ms pre-stimulus) for "hit" trials	Cohen's d = 0.95 (encoding)Cohen's d = 0.70 (baseline)	BF₁₀ = 1583.24 (encoding)BF₁₀ = 49.96 (baseline)
Generation Effect Paradigm [11]	Memory for self-generated vs. read information	• Greater pupil dilation during generation tasks• Differential effort allocation correlated with memory enhancement	Not reported	Covariance analysis showed partial mediation of generation effect
Mild Traumatic Brain Injury Assessment [12]	Neurological status (GCS scores)	• Constriction velocity correlated with GCS• Dilation velocity correlated with GCS	Strong correlation (Constriction: Sᵣ = 0.9, Dilation: Sᵣ = 0.8)	p < 0.001 (constriction)p = 0.006 (dilation)

The consistency of these findings across different memory paradigms—from free recall to recognition tasks—strengthens the evidence for pupillometry as a sensitive biomarker of memory encoding processes. The discovery that baseline pupil size (before stimulus presentation) predicts subsequent memory performance supports the "readiness to remember" (R2R) framework, suggesting that pre-stimulus attentional states significantly influence encoding success [10].

Experimental Protocols

Standardized Memory Encoding Paradigm

Table 2: Protocol for Memory Encoding with Concurrent Pupillometry

Protocol Phase	Duration	Stimuli	Measurements	Key Considerations
Participant Preparation	10 minutes	N/A	• Proper eye tracker calibration• Head stabilization	Ensure consistent lighting conditions (100-150 lux)
Baseline Recording	30 seconds	Fixation cross on neutral background	• Tonic (baseline) pupil size• Blink rate and artifacts	Use for later baseline correction of task-evoked responses
Encoding Phase	2-3 seconds per word	30 words visually presented	• Task-evoked pupil response• Absolute pupil diameter	Control for word frequency, length, and semantic properties [10]
Retention Interval	2 minutes	Distractor task (e.g., arithmetic)	N/A	Prevent rehearsal and working memory contamination
Retrieval Phase (Free Recall)	3-5 minutes	N/A	• Verbal recall recorded• Percentage of correct words	Compare pupil metrics for recalled vs. forgotten words
Retrieval Phase (Recognition)	2-3 seconds per word	30 old + 30 new words	• "Old"/"New" responses• Hit rates and false alarms	Compare pupil metrics for hits vs. misses

This protocol can be implemented using common experimental software such as E-Prime, PsychoPy, or Presentation. The retrieval phase can be adapted for either free recall or recognition tasks depending on research objectives.

Pupillometry Data Acquisition and Preprocessing

Consistent data quality is essential for reliable pupillometry results. The following standardized preprocessing pipeline is recommended based on current methodological guidelines [13]:

• Data Acquisition: Sample pupil size at ≥ 100 Hz using infrared eye-tracking systems to capture rapid pupil dynamics. The use of infrared videography provides accurate, non-invasive measurement of pupil diameter [14].

• Blink Detection and Interpolation: Identify blinks as periods of missing data or rapid pupil constriction followed by dilation. Interpolate missing data using linear or cubic spline interpolation, limiting interpolation to gaps of ≤ 150ms [13].

• Filtering: Apply a low-pass filter (e.g., Butterworth, 4 Hz cutoff) to remove high-frequency noise while preserving task-evoked responses [13].

• Baseline Correction: For trial-based analyses, subtract the mean pupil size during the 200-500ms pre-stimulus baseline from each trial to isolate phasic pupil responses [13].

• Artifact Rejection: Exclude trials with excessive blink artifacts (>30% missing data), large deviations (>3 SD from mean), or poor tracking quality.

The Scientist's Toolkit

Table 3: Essential Research Reagents and Equipment for Pupillometry Memory Research

Item	Specification	Function/Application
Infrared Eye Tracker	>100 Hz sampling rate, binocular recording	Precise measurement of pupil diameter with minimal intrusion [14]
Pupillometry Software	MATLAB/Python toolboxes with preprocessing pipelines	Data extraction, artifact removal, and feature calculation [13]
Stimulus Presentation Software	PsychoPy, E-Prime, or Presentation	Precise timing control for memory task administration
Calibration Materials	9-point calibration target, chin rest	Ensure accurate gaze tracking and minimize head movement
Standardized Word Databases	e.g., MRC Psycholinguistic Database	Control for word frequency, concreteness, emotional valence
Neutral Background Stimuli	Medium gray (RGB: 128,128,128)	Minimize light-induced pupil constriction during trials [13]

Signaling Pathways and Neurophysiological Mechanisms

The relationship between pupil responses and memory encoding is governed by well-defined neurophysiological pathways. The following diagram illustrates the primary neural mechanisms linking cognitive processing to pupillary changes:

Neural Pathways Linking Cognition to Pupil Response

This diagram illustrates the primary neural pathways through which cognitive tasks trigger pupillary changes linked to memory encoding. The sympathetic pathway (green) actively dilates the pupil via noradrenergic signaling, while parasympathetic inhibition (blue, dashed) reduces constrictor tone. These pathways converge to produce the task-evoked pupil responses that correlate with successful memory formation.

Experimental Workflow

The following diagram outlines the complete workflow for a pupillometry memory experiment, from participant preparation to data analysis:

Pupillometry Memory Experiment Workflow

This workflow encompasses the key stages of a pupillometry memory experiment. The critical preprocessing stage involves blink detection, filtering, and baseline correction to isolate task-evoked pupil responses from artifacts and intrinsic pupil fluctuations [13].

Task-evoked pupillometry provides a sensitive, non-invasive method for investigating the cognitive processes underlying memory encoding. The protocols and guidelines presented here offer researchers a standardized approach for implementing this methodology in studies of human memory. The robust relationship between pupil diameter and subsequent memory performance—evident across both free recall and recognition tasks—supports the use of pupillometry as a valuable tool for examining how attentional states and cognitive effort contribute to memory formation.

Future research directions should explore individual differences in the pupil-memory relationship, pharmacological modulation of these effects, and the application of pupillometry in clinical populations with memory impairments. The integration of pupillometry with other neuroimaging techniques, such as fMRI and EEG, will further elucidate the neural mechanisms linking pupillary responses to memory encoding processes.

Application Notes: The Inverted U-Curve in Memory Research

The inverted U-curve represents a fundamental principle in cognitive neuroscience, describing the non-linear relationship between arousal, cognitive load, and performance. This framework posits that performance, including memory formation, is optimal at intermediate levels of arousal or cognitive demand, and declines at levels that are either too low or too high [15] [16]. Pupillometry has emerged as a critical, non-invasive tool for quantifying this relationship, as task-evoked pupil dilation provides a sensitive, real-time index of cognitive effort and arousal state mediated by the locus coeruleus-norepinephrine (LC-NE) system [17] [18] [19].

The core mechanism involves the LC-NE system, where tonic (baseline) and phasic (task-evoked) activity are reflected in pupil diameter. An intermediate tonic baseline facilitates strong phasic responses to task-relevant stimuli, supporting optimal attention and memory encoding. Conversely, either low or high tonic activity results in poor phasic responses and diminished performance, creating the characteristic inverted U-shape [15] [16]. Recent research has refined our understanding, demonstrating that this relationship is not universal but is modulated by task complexity. The classic symmetric inverted U-curve is most reliably observed in tasks with simple syntactic structures, whereas it can be disrupted or become asymmetrical under conditions of high cognitive load, such as when processing complex sentence structures [20] [21].

Furthermore, the pupil-linked arousal state at the moment of encoding is a powerful predictor of long-term memory success. Studies consistently show that larger baseline and task-evoked pupil diameters during encoding are associated with subsequently recalled or recognized items, a phenomenon aligned with the "readiness to remember" (R2R) framework [5]. The table below summarizes key quantitative findings on pupillometric predictors of memory performance.

Table 1: Pupillometric Predictors of Memory Performance from Key Studies

Study	Memory Task	Pupillometric Measure	Key Finding	Effect Size (Cohen's d)
Kafkas et al. (2025) [5]	Free Recall	Baseline pupil size (500ms pre-stimulus)	Larger for recalled vs. forgotten words	0.60
Kafkas et al. (2025) [5]	Free Recall	Pupil size during encoding	Larger for recalled vs. forgotten words	0.61
Kafkas et al. (2025) [5]	Recognition	Baseline pupil size (500ms pre-stimulus)	Larger for "Hit" vs. "Miss" trials	0.70
Kafkas et al. (2025) [5]	Recognition	Pupil size during encoding	Larger for "Hit" vs. "Miss" trials	0.95
Cheng et al. (2019) [22]	Infant Visual Working Memory	Task-evoked pupil response at encoding	Correlated with subsequent memory accuracy	Not reported

The application of this knowledge is crucial for designing robust memory experiments and interpreting pupillometric data. Factors such as stimulus expectedness also modulate the pupil response; unexpected stimuli typically elicit pupil dilation predictive of subsequent memory, whereas expected stimuli can be associated with pupil constriction that also predicts memory, suggesting different underlying encoding mechanisms [19]. Therefore, controlling for task demands, stimulus properties, and baseline arousal is essential for accurately assessing cognitive load and memory formation pathways.

Experimental Protocols

The following protocols detail standardized methodologies for investigating the inverted U-curve in memory formation using pupillometry.

Protocol 1: Assessing the Inverted U-Curve with Syntactically Complex Sentences

This protocol is adapted from O'Leary et al. (2025) to examine how cognitive load, manipulated via speech rate and syntactic complexity, affects pupil dilation and recall performance [20] [21].

1. Research Reagent Solutions

Table 2: Essential Materials for Sentence Recall Studies

Item	Function	Example Specifications
Eye Tracker	Records pupil diameter at high frequency.	Tobii Pro Glasses 2 (50 Hz) or similar research-grade system [23].
Audio Presentation System	Delivers calibrated auditory stimuli.	High-fidelity headphones and sound card in a sound-attenuated booth.
Time-Compression Software	Parametrically degrades speech clarity.	Custom MATLAB or Python scripts to compress sentences to 10%, 25%, 35%, and 100% of original duration [20].
Stimulus Set	Equated sentences varying in syntactic complexity.	Subject-relative (SR) vs. Object-relative (OR) sentences matched for lexical content (e.g., "The girl that the boy sees is tall" vs. "The boy that sees the girl is tall") [21].

2. Procedure

• Participant Preparation: After obtaining informed consent, calibrate the eye tracker according to manufacturer instructions. Ensure constant ambient lighting to prevent light-induced pupil constriction [18].
• Stimulus Presentation: Present sentences auditorily in a randomized block design. Each trial consists of:
  • A fixation cross (2000 ms) to establish a pre-stimulus baseline pupil size.
  • An auditory sentence (SR or OR, presented at one of four compression rates).
  • A visual prompt for spoken recall after a short delay (e.g., "Recall now").
• Data Collection: Record continuous pupil diameter from both eyes throughout the trial. Record and score spoken recall for accuracy.
• Task Duration: Keep the total session under 60 minutes to minimize fatigue, a key confounder in pupillometry [20] [18].

3. Data Analysis

• Preprocessing:
  • Blink Correction: Interpolate missing data from blinks using a linear or cubic spline method [18].
  • Baseline Correction: For each trial, subtract the mean pupil size during the pre-stimulus baseline from the entire trial signal.
  • Filtering: Apply a low-pass filter (e.g., 4 Hz) to remove high-frequency noise [18].
• Key Dependent Variables:
  • Pupil Response: Mean pupil diameter during the sentence presentation interval, averaged across trials per condition.
  • Performance: Proportion of words correctly recalled per condition.
• Statistical Modeling: Fit the pupil response data across speech rates for each sentence type. The inverted U-curve is confirmed if a quadratic model provides a significantly better fit than a linear model for SR sentences, with a reduced or absent curve for OR sentences [20] [21].

Protocol 2: Probing "Readiness to Remember" in Long-Term Memory

This protocol is based on Kafkas et al. (2025) and is designed to investigate how pre-stimulus arousal state, indexed by baseline pupil size, predicts subsequent memory success [5].

1. Procedure

• Stimuli: A list of words (e.g., 30 nouns) presented one at a time on a neutral background.
• Encoding Phase:
  • Each trial begins with a fixation cross (2000 ms) — the critical baseline period.
  • A single word is displayed for a set duration (e.g., 3000 ms).
  • Participants make a semantic judgment (e.g., "Is this object man-made?") to ensure deep encoding.
• Retrieval Phase: After a distractor task (e.g., 10 minutes), administer either:
  • Free Recall: Participants write down all words they remember from the study phase.
  • Recognition Memory Test: Participants view old and new words and classify them as "Remember", "Know", or "New".
• Pupil Recording: Continuously record pupil size throughout both encoding and retrieval phases.

2. Data Analysis

• Trial Categorization: For the encoding phase, sort trials based on subsequent memory performance:
  • Free Recall: Recalled vs. Forgotten.
  • Recognition: Hit (Correctly identified old) vs. Miss (Incorrectly identified old).
• Baseline Pupil Size: Calculate the mean pupil diameter during the 500 ms immediately preceding word onset for each trial [5].
• Task-Evoked Pupil Response: Calculate the mean pupil diameter during the word presentation window, corrected for baseline.
• Statistical Comparison: Use paired-samples t-tests or Bayesian equivalents to compare baseline and task-evoked pupil size between subsequently remembered and forgotten items. Robust evidence is indicated by Bayes Factors (BF10) > 10 [5].

Signaling Pathways & Experimental Workflows

Neurocognitive Pathways of Pupil-Linked Memory Formation

The following diagram illustrates the principal neurobiological pathways that link cognitive demand to pupil response and memory encoding, culminating in the inverted U-curve of performance.

Diagram 1: Neurocognitive Pathways of Pupil-Linked Memory Formation

Experimental Workflow for a Pupillometry Memory Study

This workflow outlines the standard lifecycle of a pupillometry experiment, from design to analysis, highlighting critical steps for ensuring valid and reliable data.

Diagram 2: Pupillometry Experiment Workflow

The generation effect—the phenomenon where self-generated information is remembered better than passively received information—represents a cornerstone of memory research. This application note explores the pivotal role of pupillometry as an objective, physiological measure of mental effort within this phenomenon. We detail how task-evoked pupillary responses provide a reliable index of cognitive resource allocation during generative encoding, offering insights into the underlying neurobiological mechanisms. The protocols and analyses presented herein provide researchers with a framework for investigating the mental effort hypothesis of the generation effect, with direct applications in cognitive neuroscience and pharmaceutical development for cognitive disorders.

The generation effect, first formally described by Slamecka and Graf (1978), constitutes a robust memory enhancement for self-generated information compared to information that is simply read [11]. Among the several theories proposed to explain this effect, the mental effort theory posits that generation requires greater allocation of attentional resources, leading to stronger memory traces [11]. Pupillometry, the measurement of pupil diameter fluctuations, has emerged as a powerful, non-invasive tool for testing this hypothesis, as pupil dilation is a well-established correlate of mental effort, arousal, and attentional allocation [13] [24].

The physiological basis for this relationship lies within the autonomic nervous system. Pupil size is controlled by the antagonistic actions of the sphincter pupillae (constriction) and dilator pupillae (dilation) muscles. Crucially, these muscles are innervated by the parasympathetic and sympathetic nervous systems, respectively, which are influenced by central nervous system structures and neuromodulatory nuclei like the locus coeruleus-norepinephrine (LC-NE) system [24]. Activity in this system, which is integral to regulating cognitive engagement and arousal, directly causes pupil dilation, making pupillometry a sensitive proxy for the cognitive processes mobilized during effortful generation [24].

The Mental Effort Hypothesis: Evidence from Pupillometry

Early behavioral studies investigating the mental effort theory yielded conflicting results, in part due to the lack of an independent, objective measure of effort [11]. Pupillometry overcomes this limitation by providing a continuous, real-time physiological measure that is independent of task performance.

A series of four pivotal experiments using paired-associates memory tasks demonstrated that significantly more mental effort, as measured by pupil dilation, was allocated to generated information compared to read information [11]. This differential effort allocation was accompanied by a boost in memory performance, particularly in within-subjects designs. A cross-experimental covariance analysis confirmed that the allocation of differential mental effort partially accounts for the behavioral generation effect, providing strong support for the mental effort hypothesis [11].

Furthermore, converging evidence comes from research on the production effect (better memory for words read aloud vs. silently), which shares a core conceptual basis with the generation effect. Recent pupillometry studies have identified a "pupillometric production effect," where greater pupil dilation for aloud words correlates with the subsequent memory benefit [25]. This suggests that the act of production focuses attention and enhances the distinctiveness of an item's processing, mechanisms that are likely shared across various generative tasks.

Experimental Protocols for Investigating the Generation Effect

This section provides detailed methodologies for implementing pupillometry in generation effect paradigms.

Protocol 1: Paired-Associates Generation Task

This protocol is adapted from the experiments that directly tested the mental effort explanation [11].

• Objective: To determine if the act of generating a word pair member requires more mental effort than reading the pair, and if this effort predicts subsequent recall.
• Task Design:
  • Participants: A minimum of 24 participants is recommended for within-subjects designs to achieve adequate power.
  • Stimuli: A list of word pairs (e.g., "Bank - M_ _ \ Y").
  • Conditions:
    • Generate Condition: Participants are presented with a cue and a rule to generate the target word (e.g., complete the word "MONEY" using a rhyming rule).
    • Read Condition: Participants passively read the intact word pair (e.g., "Bank - Money").
  • Procedure: Trials are self-paced. Each trial begins with a fixation cross. The stimulus is then presented, and the participant either generates or reads the target. A retention interval follows, after which a memory test (e.g., cued recall) is administered.
  • Pupillometry Recording: Pupil size is recorded from the onset of the fixation cross through the response period for each trial. The key period of interest is the stimulus presentation and generation/reading phase.
• Data Analysis:
  • Preprocess pupil data to remove blinks and artifacts [13].
  • Calculate the task-evoked pupil response for each trial by subtracting the baseline pupil size (average during fixation) from the pupil size during the generation/reading phase.
  • Compare the mean pupil dilation between Generate and Read trials using paired-samples t-tests or ANOVA.
  • Conduct a covariance analysis to test if the difference in pupil dilation between conditions mediates the difference in memory accuracy.

Protocol 2: Interleaved Sustained Attention and Working Memory Task

This protocol leverages real-time triggering to link attentional states with memory performance [17].

• Objective: To probe working memory capacity at moments of high and low attentional engagement, as indexed by behavior and pupil size.
• Task Design:
  • Primary Task (Sustained Attention): Participants view displays of six shapes (circles or squares) and respond to the shape identity. One shape is highly prevalent (90%), inducing a lapsing attentional state.
  • Secondary Task (Working Memory): Participants are probed to recall the colors of the six shapes from the previous trial.
  • Real-Time Triggering:
    • Behavioral Triggering: Working memory probes are triggered when response times (RT) on the primary task are exceptionally fast (indicating a lapse) or slow (indicating an attentive state) [17].
    • Pupillometric Triggering: A novel extension involves triggering probes when real-time pupil size is exceptionally large (high arousal/effort) or small (low arousal/effort) [17].
• Data Analysis:
  • Compare working memory performance (number of colors correctly recalled) between triggered attentive and inattentive states.
  • Analyze pupil size fluctuations preceding and during the working memory probe as a function of the triggered state.

Quantitative Pupillometry Reference Data

The following table summarizes key pupillometry parameters from normative studies, which can serve as benchmarks for baseline and task-evoked responses in healthy populations.

Table 1: Normative Quantitative Pupillometry Values under Scotopic Conditions

Parameter	Mean Value (±SD)	95% Confidence Interval	Notes
Maximum Diameter (MAX)	6.6 ± 0.74 mm	5.1 – 8.1 mm	Positively correlated with age [26]
Minimum Diameter (MIN)	4.7 ± 0.77 mm	3.1 – 6.2 mm	Positively correlated with age [26]
Constriction % (CON)	30 ± 6.2%	17 – 42%	Positively correlated with age [26]
Latency (LAT)	230 ± 34 ms	160 – 300 ms	Time to initiation of constriction [26]
Average Constriction Velocity (ACV)	3.70 ± 0.74 mm/s	2.21 – 5.18 mm/s	Speed of pupil constriction [26]
Average Dilation Velocity (ADV)	0.88 ± 0.25 mm/s	0.38 – 1.38 mm/s	Speed of pupil re-dilation [26]

The Scientist's Toolkit: Essential Research Reagents and Materials

Table 2: Essential Materials for Cognitive Pupillometry Research

Item	Function & Specification	Example Use Case
Research-Grade Eye Tracker	Measures pupil diameter and gaze position at high sampling rates (≥ 120 Hz). Critical for capturing rapid task-evoked responses.	Tracking pupil dilation during a brief generation task [11] [5].
Stimulus Presentation Software	Software capable of precise timing and synchronization with the eye tracker (e.g., PsychoPy, E-Prime).	Presenting generation task stimuli and marking trial events in the pupil data stream [17] [13].
Data Preprocessing Pipeline	Custom or commercial software for parsing, cleaning, and analyzing pupil data. Handles blink removal, smoothing, and baseline correction.	Isolating the task-evoked pupil response from the raw pupil size signal [13].
Chromatic-Calibrated Display	A monitor with controlled and calibrated color and luminance output. Essential for controlling for the pupillary light reflex.	Ensuring stimuli across 'generate' and 'read' conditions are photometrically matched [13] [27].
Far-Red Filter System (Prototype)	A long-pass filter (~630 nm) attached to a standard RGB camera. Increases pupil-iris contrast, especially for dark irises, improving accessibility.	Enabling pupillometry with commodity hardware (e.g., smartphones) for broader deployment [28].

Signaling Pathways and Experimental Workflows

Neurobiological Pathways of Pupil Control

The following diagram illustrates the autonomic nervous system pathways that control pupil size and their link to cognitive processes involved in the generation effect.

Experimental Workflow for a Generation Effect Study

This workflow outlines the step-by-step process for conducting a pupillometry study on the generation effect, from setup to data interpretation.

Pupillometry provides a robust, objective, and non-invasive method for quantifying the mental effort underlying the generation effect. The protocols and guidelines detailed in this document empower researchers to rigorously test cognitive theories and investigate the neurocognitive mechanisms of memory enhancement. In translational and pharmaceutical contexts, this paradigm offers a powerful tool for assessing the efficacy of cognitive enhancers or neuromodulatory drugs by measuring their impact on cognitive effort allocation during encoding, providing a sensitive biomarker for intervention studies in populations with cognitive deficits.

A Practical Framework for Designing and Executing Pupillometry Studies on Memory

Pupillometry, the measurement of pupil diameter and reactivity, has emerged as a powerful, non-invasive tool for investigating cognitive processes. This application note details core trial-based experimental designs that utilize pupillometry to study the intricate relationship between attention and working memory. The pupil provides a unique window into cognitive and autonomic nervous system activity, with its size fluctuating in response to factors beyond ambient light, including cognitive load, arousal, and the strength of memory signals [29] [17]. These fluctuations are linked to the activity of the locus coeruleus-norepinephrine (LC-NE) system, which regulates arousal and attention [29] [30]. The protocols outlined herein are designed for researchers and scientists in both academic and drug development settings, providing robust frameworks for quantifying cognitive states and assessing the efficacy of pharmacological or other interventions.

Physiological and Neural Basis of Cognitive Pupillometry

Pupil size is controlled by the antagonistic actions of two muscles: the iris sphincter (constricting) and the iris dilator (dilating). These are governed by distinct neural pathways [29].

• The Parasympathetic Constriction Pathway: This pathway mediates the pupil light reflex (PLR). Light information travels from the retina via the pretectal nucleus to the Edinger-Westphal nucleus (EWN), and then through the ciliary ganglion to the iris sphincter muscle, causing constriction [29].
• The Sympathetic Dilation Pathway: Pupil dilation is tied to arousal and cognitive effort. The locus coeruleus (LC) and hypothalamus project to the intermediolateral column of the spinal cord, which then connects to the superior cervical ganglion. This pathway ultimately innervates the iris dilator muscle. Crucially, the LC also inhibits the EWN, providing a mechanism for cognitive arousal to induce dilation by suppressing the constriction pathway [29].

The following diagram illustrates these interacting pathways:

Diagram 1: Neural pathways controlling pupil size. The parasympathetic pathway (green) drives constriction, while the sympathetic pathway (red) drives dilation. A key interaction is the inhibitory projection from the LC to the EWN, allowing arousal to cause pupil dilation [29].

Core Experimental Paradigms

This section provides detailed protocols for two primary trial-based paradigms that interleave pupillometry with tasks probing attention and memory.

Interleaved Sustained Attention and Working Memory Task

This paradigm examines the synchronous fluctuations of attention and working memory capacity within a single task [17].

3.1.1 Experimental Workflow

The following diagram outlines the structure of a single trial and the real-time triggering logic:

Diagram 2: Workflow of the interleaved attention and working memory task, featuring real-time behavioral triggering [17].

3.1.2 Detailed Protocol

• Participants: 20-35 participants per group to achieve sufficient statistical power [17].
• Apparatus: An eye tracker (e.g., EyeLink 1000 Plus) recording at 500 Hz, a chin-rest to stabilize head position, and a monitor positioned approximately 70 cm from the participant [31] [17].
• Stimuli:
  • Sustained Attention Task: Displays of six circles or squares are presented. The task-relevant dimension is shape.
  • Working Memory Task: The task-relevant dimension is the color of each shape (e.g., red, blue, green, etc.). This orthogonality ensures the tasks are distinct [17].
• Procedure: Participants complete multiple blocks (e.g., 5 blocks of 800 trials). On each trial:
  • A fixation dot is presented.
  • The stimulus display (six shapes) is presented for 800 ms.
  • The participant indicates if all shapes are circles or squares via keypress.
  • The screen blanks until the next trial.
• Real-Time Triggering: Working memory probes are not random. After an initial set of trials, the system calculates a running average of reaction times (RTs). When a participant's RT deviates significantly (e.g., faster or slower than one standard deviation from their mean), it indicates a lapsing or highly attentive state, respectively, and triggers a working memory probe [17].
• Working Memory Probe: After a 2-second delay, multicolored squares appear at all previous locations. Participants must use a mouse to select the color that was originally presented at each location.
• Pupillometry Measures: Pupil size is recorded continuously throughout the task. Key metrics include baseline pupil size before stimulus onset, mean pupil diameter during the retention interval, and task-evoked pupil dilation in response to stimuli [17].

Continuous Recognition Memory Paradigm

This paradigm investigates how memory strength and subjective familiarity are reflected in pupil activity by manipulating the retention interval between initial exposure and memory test [30].

3.2.1 Experimental Workflow

The following diagram illustrates the trial structure across different retention intervals (lags):

Diagram 3: Workflow of the continuous recognition memory task with multiple retention intervals (lags) to manipulate memory strength [30].

3.2.2 Detailed Protocol

• Participants: Approximately 40 participants to account for potential data loss, ensuring a final sample of >35 [30].
• Stimuli: A large set of words (e.g., 100+), matched for psycholinguistic variables like familiarity and word length [30].
• Design: A within-subjects design where each participant is tested at multiple retention intervals (lags), such as 1, 4, 8, and 32 intervening items between the study and test presentation of a word [30].
• Procedure:
  • Study Phase: Each trial begins with a fixation cross (250 ms), followed by a word presented for 1750 ms. Participants are instructed to remember the word.
  • Filler Items: A sequence of words is presented, including both new words and words that will be tested later. The number of these intervening words determines the lag condition.
  • Test Phase: A mask (e.g., "&&&&&&&") is presented for 250 ms, followed by a word (1750 ms). Participants must indicate if the word is "old" (seen before in the experiment) or "new" (seen for the first time). Following this, they provide a confidence rating for their decision [30].
• Pupillometry Measures: Pupil size is recorded time-locked to the test stimulus onset. The analysis focuses on the "pupil old/new effect"—the difference in pupil dilation between correctly identified old words ("hits") and correctly rejected new words ("correct rejections"). The time course of the pupillary response is also analyzed, as early and late components may relate to objective memory strength and subjective familiarity, respectively [30].

Data Presentation and Analysis

Quantitative Pupillometry Reference Values

The following table summarizes key pupillometry parameters and their normative values, which can serve as a baseline for evaluating task-induced changes or pathological deviations.

Table 1: Normative Quantitative Pupillometry Values in a Pediatric Population (Scotopic Conditions) [26] [32]

Parameter	Description	Mean ± SD	95% CI
MAX (mm)	Maximum (resting) pupil diameter	6.6 ± 0.74 mm	5.1 – 8.1 mm
MIN (mm)	Minimum (constricted) diameter	4.7 ± 0.77 mm	3.1 – 6.2 mm
CON (%)	Constriction percentage	30 ± 6.2 %	17 – 42 %
LAT (ms)	Latency to start of constriction	230 ± 34 ms	160 – 300 ms
ACV (mm/s)	Average constriction velocity	3.70 ± 0.74 mm/s	2.21 – 5.18 mm/s
MCV (mm/s)	Maximum constriction velocity	5.02 ± 0.90 mm/s	3.22 – 6.82 mm/s
ADV (mm/s)	Average dilation velocity	0.88 ± 0.25 mm/s	0.38 – 1.38 mm/s

Expected Behavioral and Pupillometric Outcomes

Table 2: Typical Results from Core Cognitive Paradigms

Paradigm	Behavioral Outcome	Pupillometry Outcome	Cognitive Interpretation
Interleaved Attention/WM [17]	Faster RTs indicate lapsing attention; slower RTs indicate high attention. WM performance is worse on fast-RT trials.	Larger baseline pupil size and greater task-evoked dilation during high-attention (slow RT) states.	Pupil size covaries with attentional state and working memory load. Larger pupils index greater cognitive engagement and arousal.
Continuous Recognition Memory [30]	Recognition accuracy decreases as retention interval (lag) increases.	The "pupil old/new effect" (dilation for hits vs. correct rejections) is strongest at short lags and diminishes with longer lags.	Pupil dilation at retrieval is associated with the strength of the memory signal. Weaker memories (long lag) elicit a reduced pupillary response.
Stimulus Orientation (Control) [31]	No behavioral difference due to task-irrelevant stimulus feature.	Systematically smaller pupils for vertical vs. horizontal lines, despite perfect luminance matching.	Pupil is sensitive to subtle, low-level visual features and cognitive states unrelated to luminance, highlighting a potential confound.

The Scientist's Toolkit

Table 3: Essential Research Reagent Solutions and Materials

Item	Function / Application	Example Specifications
Research-Grade Eye Tracker	Precisely records pupil diameter and gaze position at high frequencies.	- High-Precision: EyeLink 1000 Plus (500 Hz+) [31].- Portable/Wearable: Pupil Labs Core or Invisible [33].
Stimulus Presentation Software	Precisely controls the timing and presentation of experimental paradigms.	PsychoPy (Python-based) [31] [17].
Pupillometry Data Processing Software	Processes raw eye-tracking data to extract pupil size, removes artifacts (blinks), and calculates metrics.	- Pupil Labs Suite: Pupil Cloud & SPEED software [33].- Custom Scripts: (e.g., in Python or MATLAB).
Quantitative Pupillometer (Clinical)	Provides a standardized, handheld method for measuring the Pupillary Light Reflex (PLR) in clinical settings.	NeurOptics PLR-200 pupillometer [26].
Standardized Stimulus Sets	Ensures experimental consistency and replicability in memory and emotion research.	- Words: Databases with normative ratings (familiarity, arousal) [30].- Faces: Standardized sets of emotional faces (e.g., for attachment/emotion studies) [34].
Chin/Forehead Rest	Minimizes head movement, ensuring stable and clean pupil recordings.	Adjustable, comfortable rest for a 70-75 cm viewing distance [31] [17].

Critical Methodological Considerations

• Luminance Matching is Not Sufficient: A fundamental finding is that even perfectly luminance-matched stimuli can produce systematic pupil responses if they differ in other features, such as orientation (vertical vs. horizontal). This "orientation effect" underscores the need to control for non-luminance features in cognitive pupillometry [31].
• Real-Time Triggering: Using behavior (e.g., RT) or physiology (e.g., pupil size) to trigger experimental events is a powerful method for probing specific cognitive states (e.g., attentional lapses) as they occur naturally, increasing the validity and power of the experiment [17].
• Baseline Correction: Pupil responses are slow and additive. It is critical to define a pre-stimulus baseline period (e.g., 200-500 ms before stimulus onset) and subtract this baseline from the trial data to isolate task-evoked pupil responses from slow drifts and overall arousal level [30].

Pupillometry, the measurement of pupil dilation, has become an established method in psychological and psycholinguistic research for tracing online cognitive and emotional processing [35]. The pupil's response is a highly sensitive measure of physiological arousal and cognitive effort, modulated by the autonomic nervous system [34] [36]. However, this sensitivity is also a significant drawback: the pupillary signal is influenced by numerous confounding factors unrelated to the experimental manipulation [35]. Changes in pupil size occur involuntarily as a result of low-level (e.g., light reflex), mid-level (e.g., alertness), and high-level processes (e.g., executive functioning) [34]. Therefore, to ensure that observed differences in pupil dilation genuinely reflect the cognitive or emotional processes of interest—such as attention in memory research—researchers must rigorously control for potential confounds. This document outlines detailed protocols for controlling three primary categories of confounds: luminance, stimulus properties, and participant factors, specifically within the context of memory research.

Luminance Control Protocols

Experimental Design and Setup

Luminance is the most critical low-level confound in pupillometry research. Any variation in luminance across experimental conditions will trigger the pupil light reflex, which can easily overshadow cognitively-driven pupillary responses.

• Constant Illumination: Conduct experiments in a dimly lit, windowless room where all ambient lighting can be precisely controlled. Use neutral, uniform background colors on the display monitor, typically a mid-level gray (e.g., RGB: 128, 128, 128) to minimize reflections and maintain consistent baseline adaptation [35].
• Stimulus Luminance Matching: For all visual stimuli (e.g., images, words, faces), luminance values must be mathematically equated across experimental conditions. Use software such as MATLAB or Python with the Psychtoolbox to calculate and verify the mean luminance and root mean square (RMS) contrast of every image. For facial stimuli, a common approach is to convert images to grayscale and match histograms.
• Apparatus Calibration: Use a photometer to regularly calibrate the display monitor to ensure linear gamma correction and consistent luminance output across the entire screen.

Table 1: Quantitative Luminance and Color Values for Standardized Backgrounds

Background Type	Hex Code	RGB Values	Luminance (cd/m²)	Use Case
Mid-Gray	#808080	128, 128, 128	~50	Standard background for most cognitive tasks
White	#FFFFFF	255, 255, 255	~120	High-luminance control
Black	#000000	0, 0, 0	~0	Low-luminance control

Workflow for Luminance Control

The following diagram illustrates the standardized protocol for ensuring luminance control throughout a pupillometry experiment.

Controlling for Stimulus Properties

Design and Norming Procedures

In memory research, the properties of the to-be-remembered stimuli can systematically influence pupillary responses. Controlling for these is essential for clean interpretation.

• Stimulus Selection and Norming: For verbal memory tasks, select words from standardized databases that control for frequency, concreteness, age of acquisition, word length, and emotional valence [35] [36]. For instance, when investigating the cognitive effort of word recognition, Kuchinke et al. (2007) controlled for word frequency, and Mathôt et al. (2017) controlled for semantic brightness [36].
• Use of Control Conditions: Employ well-matched control conditions to isolate specific processes. For example, to study memory encoding, compare pupil dilation to items that are later remembered versus those that are later forgotten. To study the effort of processing novelty, compare familiar real words (e.g., "table") to unfamiliar pseudowords (e.g., "flirp"), ensuring length and pronounceability are matched [36].
• Experimental Design Considerations: Counterbalance or randomize the presentation of stimuli across participants to avoid confounding stimulus properties with time-based effects like fatigue or practice. For studies with trial-unique feedback images (e.g., to measure attention to reinforcement during memory tasks), ensure the images are pre-tested for emotional content, complexity, and luminance [37].

Table 2: Key Stimulus Properties to Control in Memory Research

Stimulus Property	Description	Measurement/Control Method	Impact on Pupil Dilation
Word Frequency	How often a word appears in a language corpus.	Use standardized frequency norms (e.g., SUBTLEX).	Lower frequency words elicit greater dilation, indicating more effort [36].
Emotional Valence	The positive or negative character of a stimulus.	Use norms from databases (e.g., IAPS, ANEW).	Negative stimuli often elicit larger dilation than positive or neutral ones [34].
Word Length	Number of letters or syllables.	Match across conditions (e.g., real words vs. pseudowords).	Longer words can cause greater dilation, especially for unfamiliar items [36].
Visual Complexity	Amount of detail or clutter in an image.	Compute RMS contrast or use subjective ratings.	High complexity can increase processing load and pupil dilation.
Arousal Level	The intensity of emotion evoked.	Use normative ratings from databases.	Higher arousal stimuli lead to greater sympathetic activation and pupil dilation [34].

Workflow for Managing Stimulus Properties

The following protocol ensures that stimulus properties are adequately controlled from the design phase through to data analysis.

Accounting for Participant Factors

Screening, Assessment, and Experimental Design

Individual differences between participants can be a major source of variance in pupillary responses and must be measured and controlled.

• Baseline Pupil Size: Baseline pupil size is influenced by age, fatigue, caffeine intake, and time of day. Protocol: Implement a pre-experiment fixation period (e.g., 30-60 seconds) where participants view a neutral screen. Calculate the mean pupil size during this period for use as a covariate in statistical models. This helps control for tonic (baseline) arousal levels.
• Age and Sex: Pupil size and responsiveness decrease with age. Protocol: Record participant age and sex. Either match participant groups on these variables or include them as covariates in linear mixed models [34].
• Trait-Level Individual Differences: Certain traits are known to modulate physiological responses to emotional or cognitively demanding stimuli. Protocol: Administer standardized self-report questionnaires. For example, in emotion processing studies, adult attachment styles (e.g., dismissing orientation) can modulate pupil dilation to negative stimuli [34]. In motivation research, sensitivity to reward and punishment can be assessed [37].

Table 3: Essential Participant Factors and Control Methods

Participant Factor	Impact on Pupillometry	Control Method
Baseline Arousal	Influences starting point for task-evoked dilation; correlates with fatigue, caffeine, etc.	Pre-trial baseline correction; fixed resting baseline period.
Age	Reduces maximum pupil size (senile miosis) and constriction velocity.	Statistically covary; use age-matched participant groups.
Sex	May interact with emotional processing; some studies show no direct main effect [34].	Record and include as a fixed factor in statistical models.
Trait Affect/Personality	e.g., Dismissing attachment heightens arousal to negative emotions [34]; reward sensitivity affects motivation.	Administer relevant questionnaires (e.g., ECR, SPSRQI) and use scores as covariates or moderators.

The Scientist's Toolkit: Research Reagent Solutions

Table 4: Essential Materials and Reagents for Pupillometry Research

Item Name	Function/Application	Example Specifications
Eye-Tracker	Records pupil diameter and gaze position at high temporal resolution.	Model: EyeLink 1000 Plus; Sampling Rate: 1000 Hz; Accuracy: < 0.5°.
Stimulus Presentation Software	Prescribes the experimental paradigm; precise timing of stimulus onset is critical.	Software: PsychoPy, E-Prime, or MATLAB with Psychtoolbox.
Standardized Stimulus Sets	Provides pre-normed stimuli to control for confounding properties.	Databases: International Affective Picture System (IAPS), Affective Norms for English Words (ANEW), MRC Psycholinguistic Database.
Pupillometry Analysis Software	Preprocesses raw pupil data (filtering, blink correction) and extracts metrics.	Tools: Pupil Labs, EyeLink software, or custom scripts in R/Python.
Linear Mixed Models (LMM) Software	Statistically models time-series pupillometry data while accounting for random effects and autocorrelation.	Platform: R with packages lme4 for LMM and mgcv for GAMMs [35].
Psychometric Questionnaires	Quantifies participant traits that may confound or moderate pupillary responses.	Examples: Experiences in Close Relationships (ECR) for attachment [34], Sensitivity to Punishment and Reward Questionnaire (SPSRQI) [37].

Integrated Experimental Protocol for Memory Research

The following integrated workflow combines the control of luminance, stimulus properties, and participant factors into a single protocol for a pupillometry study on memory encoding.

Statistical Analysis and Data Modeling

Addressing Time-Series Specific Challenges

Pupillometry data presents unique statistical challenges that must be addressed to avoid false positives.

• Analyzing the Full Time Course: Instead of extracting single features (e.g., peak dilation), analyze the entire pupil dilation trajectory using methods like Generalized Additive Mixed Models (GAMMs) or growth curve analysis with linear mixed-effects models. This approach utilizes all available information and provides a more coherent interpretation of when conditions differ over time [35].
• Accounting for Autocorrelation: Pupillary signals are inherently autocorrelated, meaning data points close in time are more similar. This extreme autocorrelation increases the probability of Type I errors. GAMMs offer the possibility to include an autoregressive (AR) error model to deal with this structure directly [35].
• Incorporating Random Effects: Use linear mixed models with random intercepts for participants and items (stimuli) to account for the large within- and between-subject variability inherent in pupil dilation recordings. This prevents conclusions from being anticonservative [35].

A sample statistical model for analyzing the effect of a memory condition on pupil dilation might be specified in R as follows, controlling for key confounds: lmer(PupilSize ~ Condition * Time + BaselinePupil + WordFrequency + Age + (1|Participant) + (1|Stimulus), data = my_data)

Pupillometry has emerged as a powerful, non-invasive tool for investigating cognitive processes, including attentional mechanisms in memory formation. This application note details the essential hardware and software specifications for implementing pupillometry in memory research, with a specific focus on capturing the physiological correlates of attention during encoding and retrieval processes. The measurement of pupil diameter provides a unique window into cognitive effort, arousal, and the activity of the locus coeruleus-norepinephrine (LC-NE) system, which are all critical for understanding memory [5] [38]. Proper data acquisition is paramount, as the pupil response is a delicate physiological signal influenced by numerous external and internal factors. Adherence to rigorous hardware and software standards ensures the reliability and validity of the data, enabling researchers to draw meaningful conclusions about the interplay between attention and memory.

Essential Hardware Specifications

The core of a pupillometry setup is the eye tracker, which must meet specific technical criteria to capture the subtle, rapid changes in pupil size that reflect cognitive processing.

Core Pupillometry Hardware

Component	Minimum Specification	Rationale
Eye Tracker	Monocular or binocular sampling rate ≥ 60 Hz (≥ 120 Hz recommended for dynamics)	Adequate temporal resolution is required to capture the rapid, task-evoked pupil response [39].
Camera Type	Infrared-sensitive camera	Illuminates the pupil with invisible IR light and captures high-contrast images for accurate pupil detection [40].
Spatial Resolution	Sufficient to detect pupil size changes with an accuracy of < 0.1 mm	High precision is needed to measure small but statistically significant dilations related to cognitive load [5].
Light Source	Controlled, constant infrared illumination	Provides consistent lighting for the camera without influencing the participant's pupil response [13].
Stimulus Display	Monitor with precise timing (refresh rate ≥ 60 Hz)	Ensures accurate presentation and timing of experimental stimuli [39].
Data Synchronization	Hardware or software sync between eye tracker, display, and response input	Critical for aligning pupil data with specific experimental events (e.g., stimulus onset) [13].

Experimental Control and Environment

Controlling the experimental environment is as crucial as the hardware itself. Key considerations include:

• Stimulus Control: Visual stimuli presented during memory tasks (e.g., words, images) must be carefully matched for average luminance, contrast, and color across experimental conditions. Even when brightness is identical, low-level features like color and visual change can induce pupil constriction and confound results [13] [27].
• Lighting: Testing must be conducted in a room with stable, dim ambient lighting. Scotopic (dark) or mesopic (dim) conditions are typically used to allow for observable pupil dilation, and lighting levels must be verified with a luminometer and reported in the methodology [41].
• Participant Stability: Use a chin rest or forehead support to minimize head movements and reduce artifacts in the pupil recording [39].

Essential Software and Data Processing Specifications

Software is required for stimulus presentation, pupil data extraction, and preprocessing to clean the raw signal before analysis.

Software Components and Functions

Software Component	Core Function	Key Features & Outputs
Stimulus Presentation	Presents experimental protocol and records behavioral responses.	Precise timing, synchronization marks sent to eye tracker, compatibility with common psychology software (e.g., PsychoPy, E-Prime).
Pupil Detection Algorithm	Processes video stream to identify pupil and calculate diameter.	Sub-pixel accuracy, robust to partial occlusions, outputs pupil size (in mm or pixels) and confidence metrics over time [40].
Data Preprocessing Toolbox	Cleans raw pupil data by identifying and correcting artifacts.	Automated blink detection and interpolation (e.g., linear or cubic-spline); algorithms based on velocity changes or missing data [13] [40].
Data Analysis Platform	Extracts outcome measures and performs statistical tests.	Capability to calculate baseline pupil size, task-evoked pupil response (TEPR), constriction/dilation velocity, and post-illumination pupil response (PIPR) [13] [40].

Preprocessing Workflow

A standardized preprocessing workflow is vital for data quality. The following diagram outlines the key stages, from raw data to a clean, analysis-ready signal.

Figure 1: Pupil Data Preprocessing Workflow.

The workflow involves several critical steps:

• Filtering: Applying a low-pass filter to remove high-frequency noise from the raw signal [40].
• Artifact Detection: Identifying periods of signal loss or corruption, most commonly caused by blinks, using methods like velocity threshold analysis or the detection of missing data [13] [40].
• Interpolation: Replacing the identified artifact periods with estimated values using linear or cubic-spline interpolation to create a continuous signal [40].
• Baseline Correction: For trial-based designs, pupil size during a pre-stimulus baseline period is subtracted from the signal to isolate task-evoked changes [5] [13].

Specialized software like PupilMetrics can automate much of this workflow, significantly reducing processing time while maintaining accuracy comparable to manual analysis [40].

Experimental Protocol: Predicting Long-Term Memory

To illustrate the application of these specifications, we detail a protocol from a recent study that used pupillometry to predict long-term memory formation based on pre-stimulus arousal [5].

Experimental Design and Parameters

This protocol is designed to investigate the "readiness to remember" (R2R) framework.

• Objective: To determine if baseline pupil size (a measure of pre-stimulus arousal) and pupil size during encoding predict subsequent free recall and recognition memory performance.
• Participants: Ninety-five psychology students.
• Stimuli: A list of words. The specific words should be controlled for factors like length, frequency, and emotional valence.
• Design: Within-subjects, where each participant's performance on "recalled" vs. "forgotten" trials is compared.
• Procedure:
  • Baseline Period: A fixation cross is presented for a set duration (e.g., 1000 ms) before each word. Pupil size is measured during the final 500 ms of this period [5].
  • Encoding Phase: A single word is presented on the screen for several seconds (e.g., 3000 ms) while the participant's pupil is tracked.
  • Distractor Task: A brief mathematical task or similar activity to prevent rehearsal.
  • Retrieval Phase:
    • Experiment 1 (Free Recall): Participants verbally report as many words as they can remember.
    • Experiment 2 (Recognition): Participants are presented with "old" (studied) and "new" (unstudied) words and must classify them.
• Key Pupillometry Outcomes:
  • Baseline Pupil Size: Average diameter in the 500 ms before word onset.
  • Encoding Pupil Size: Average diameter during the entire word presentation period.

Data Acquisition and Analysis Specifications

The following table outlines the critical technical parameters for implementing this protocol.

Parameter	Specification for Memory Protocol
Sampling Rate	250 Hz (or higher) to capture rapid fluctuations.
Baseline Period	500 ms pre-stimulus, measured immediately before word onset [5].
Trial Structure	Fixed inter-trial interval to allow pupil return to baseline.
Data Extraction	Pupil size time-locked to word presentation, averaged per trial for baseline and encoding periods.
Statistical Analysis	Comparison of mean pupil size (baseline and encoding) between later "recalled" and "forgotten" items using Bayesian or repeated-measures ANOVA [5].

The Scientist's Toolkit: Key Research Reagents & Materials

This section catalogs the essential "research reagents" and materials required to conduct a pupillometry study in memory research.

Item	Function / Relevance
Infrared Eye Tracker	The primary data collection device. Must provide the necessary spatial and temporal resolution for cognitive phenomena [39].
Stimulus Presentation Software	Software like PsychoPy or E-Prime to display memory tasks with millisecond precision and send synchronization pulses [13].
Quantitative Pupillometry Analysis Toolbox	Software libraries (e.g., in Python, R, or MATLAB) for preprocessing and analyzing pupil data, such as the PupilMetrics platform [13] [40].
Chin/Forehead Rest	Stabilizes the participant's head to minimize motion artifacts and ensure consistent data quality [39].
Luminometer	Measures and verifies ambient light levels in the testing environment to ensure consistency across participants and sessions [41].
Validated Memory Stimuli Sets	Word lists or image sets (e.g., IAPS) normed for emotionality, frequency, and complexity, and critically, matched for low-level visual features like luminance [5] [13].

Integrated Pathway: From Stimulus to Memory Prediction

The following diagram synthesizes the physiological pathway, experimental workflow, and logical relationships that underpin the use of pupillometry in memory research, as demonstrated in the featured protocol.

Figure 2: Physiological and Experimental Pathway.

This integrated pathway shows how the experimental measure (pupil size) is linked to a underlying neurophysiological system. Activity in the Locus Coeruleus (LC) triggers the release of Norepinephrine (NE), which causes pupil dilation and enhances attention and arousal. This state of heightened attention facilitates more effective memory encoding, leading to successful long-term memory formation. The pupillometry measurement of baseline and encoding-phase pupil diameter serves as a non-invasive proxy for this entire pathway, allowing researchers to predict subsequent memory performance based on a participant's physiological state before and during learning [5] [38].

Pupillometry provides a non-invasive window into cognitive processes such as attention, arousal, and mental effort, making it particularly valuable in memory research where it can serve as a psychophysiological readout of autonomic and cognitive processes [42]. The pupil's diameter is regulated by the sympathetic nervous system and is mediated primarily via norepinephrine from the locus coeruleus (LC), establishing it as a reliable measure of LC-noradrenergic activity under constant lighting conditions [43]. In memory studies, pupil dilation has been linked to encoding, retrieval, and working memory operations, offering insights into the cognitive load and resource allocation during these processes [42].

However, raw pupillometry data contains substantial noise from various sources, including blinks, head movements, and measurement artifacts [43]. Preprocessing is therefore a fundamental prerequisite for high-quality data analysis, transforming raw, noisy recordings into a clean, interpretable time series [42] [43]. This Application Note provides a standardized, step-by-step protocol for preprocessing pupil diameter data, specifically contextualized for research investigating attentional processes in memory.

The Pupillary Cognitive Pathway: A Neurological Framework

Pupil size changes associated with cognition are primarily governed by a specific neural pathway. The diagram below illustrates this pathway, from stimulus perception to the final pupillary response.

Diagram 1: The primary neural pathway mediating cognitive pupillary responses. Under constant luminance, cognitive stimuli activate the Locus Coeruleus (LC), leading to Norepinephrine (NA) release. NA acts on α-adrenoceptors in the iris dilator muscle, causing pupil dilation. This dilation is measured as a non-invasive surrogate of LC-NA activity, providing a readout of cognitive processes like attention and mental effort [42] [43].

Experimental Protocol: A Standardized Preprocessing Pipeline

The following protocol is designed to handle data from common eye-tracking systems and prepares it for statistical analysis, including advanced time-series approaches.

Materials and Data Acquisition

Table 1: Essential Research Reagents and Solutions

Item	Function/Specification	Application Notes
Eye-Tracker (e.g., Smart Eye Aurora)	Records raw pupil coordinates and diameter at high frequencies (e.g., 60-120 Hz).	Ensure proper calibration. Binocular recording is recommended for data redundancy [44].
Data Acquisition Software (e.g., iMotions)	Synchronizes pupillometry with other data streams (e.g., facial video, EEG).	Critical for multimodal studies of attention and memory [44].
Stimulus Presentation Software	Preserts memory tasks (e.g., item recognition, serial recall).	Must output event markers (triggers) synchronized with the eye-tracker.
Processing Environment (e.g., MATLAB, R, Python)	Implements the preprocessing code.	Tools like eyeris (Python) provide a standardized, FAIR-compliant framework [42].

Step-by-Step Preprocessing Workflow

The entire pipeline, from raw data to a clean, analysis-ready signal, is summarized in the following workflow diagram.

Diagram 2: A comprehensive preprocessing workflow for pupillometry data. The pipeline involves preparing raw data, detecting and removing multiple types of artifacts, interpolating missing data, smoothing the signal, and applying baseline correction to produce a clean time series for statistical analysis [43] [44].

Step 1: Data Preparation and Segmentation

• Action: Convert proprietary eyetracker output into a standard format containing a continuous pupil diameter time series and trial segmentation information based on experimental event markers [43].
• Protocol:
  • Import data for left and right eyes separately.
  • Extract and align task event markers (e.g., stimulus onset, response) with the pupil data.
  • Remove non-positive values or samples explicitly marked as invalid by the eyetracker.
• Output: A standardized data matrix with columns for timestamp, left pupil diameter, right pupil diameter, and trial markers.

Step 2: Artifact Detection and Filtering

This critical step removes biologically implausible data points. The following table summarizes key artifact types and their detection methods.

Table 2: Artifact Detection Methods and Thresholds

Artifact Type	Detection Method	Typical Threshold	Rationale
Invalid Range [43]	Absolute value check.	Remove values < 2 mm or > 8 mm.	Values outside this range are physiologically implausible in humans and indicate tracking loss.
Dilation Speed Outliers [43] [44]	Median Absolute Deviation (MAD) of normalized dilation speed.	Threshold = Median + (k * MAD). Common k: 2 to 3 [44].	Identifies unrealistically fast changes caused by blinks or measurement errors. Normalized speed = |d[i] - d[i-1]| / (t[i] - t[i-1]).
Temporally Isolated Samples [43]	Check the number of consecutive valid samples.	Remove valid samples in a run of < 3 [43].	Isolated valid samples surrounded by invalid data are likely noise and should not be interpolated.

Step 3: Interpolation

• Action: Fill gaps created by artifact removal to create a continuous signal.
• Protocol: Use the Piecewise Cubic Hermite Interpolating Polynomial (PCHIP) method to interpolate over short gaps [44]. PCHIP is preferred over linear or spline interpolation as it avoids creating spurious peaks, which is crucial for analyzing phasic pupil responses in memory tasks.
• Note: Interpolate only over short gaps (e.g., < 150 ms). Longer gaps should be excluded from analysis.

Step 4: Smoothing and Filtering

• Action: Attenuate high-frequency noise that is not physiologically relevant.
• Protocol: Apply a low-pass filter (e.g., a Butterworth filter) with a cut-off frequency of 4-6 Hz [43]. This preserves the signal components related to cognitive activity while removing high-frequency noise. Alternatively, a moving average filter can be used for simplicity.

Step 5: Baseline Correction and Output

• Action: Account for individual differences and slow drifts in tonic pupil size by referencing the signal to a pre-stimulus baseline period.
• Protocol:
  • For each trial, calculate the mean pupil diameter during a baseline period (e.g., 200-500 ms before stimulus onset).
  • Subtract this baseline value from every data point in the trial epoch.
  • The output is a clean, baseline-corrected pupil diameter time series for each trial, ready for statistical analysis [43].

Statistical Considerations for Memory Research

Analyzing the full time course of the pupil dilation signal, rather than just extracted features (e.g., mean or peak dilation), utilizes all available information and can reveal the precise timing of cognitive events [35]. Generalized Additive Mixed Models (GAMMs) are particularly suited for this, as they can handle the nonlinear shape of pupil responses and account for autocorrelation and random effects of participants and items [35].

A significant challenge in time-series analysis is autocorrelation, where residuals from adjacent time points are correlated, inflating the risk of Type I errors (false positives) [35]. GAMMs can incorporate an autoregressive (AR) error model to address this issue, which is critical for producing valid statistical inferences in pupillometry research [35].

This Application Note outlines a robust and standardized preprocessing workflow essential for producing reliable, high-fidelity pupillometry data. In memory research, where pupil dynamics can reflect subtle cognitive processes related to attention and encoding, rigorous preprocessing is not merely a technical step but a foundational component of experimental validity. Adopting standardized pipelines like eyeris or the protocol described here enhances the reproducibility and comparability of findings across studies, ultimately strengthening the use of pupillometry as a core tool in cognitive neuroscience [42].

Pupillometry, the measurement of pupil diameter fluctuations, has become an essential tool in cognitive neuroscience for studying attention, memory, and autonomic nervous system activity. Within memory research, pupil dynamics provide a non-invasive window into cognitive processes during encoding, maintenance, and retrieval. The analytical approaches of baseline correction, normalization, and statistical modeling are fundamental for extracting meaningful cognitive indices from pupillary data, particularly in research focusing on attention-mediated memory formation. These methods enable researchers to isolate task-evoked cognitive responses from physiological noise and individual differences, thereby revealing neural mechanisms underlying memory performance. This application note details established protocols and analytical frameworks for pupillometry data processing, with specific application to attention and memory research.

Core Analytical Frameworks

Baseline Correction

Baseline correction is a critical first step in pupillometry analysis that isolates phasic, task-evoked pupil responses from tonic, pre-stimulus pupil diameter. This process accounts for fluctuating arousal states and establishes a reference point for quantifying stimulus-related changes.

Physiological Rationale: Tonic (baseline) pupil size reflects the relative balance of the autonomic nervous system and is correlated with the baseline firing rate of the locus coeruleus (LC), a key neural structure for attention and memory [10] [45]. Phasic pupil responses, measured from this baseline, are linked to the phasic firing of the LC-noradrenergic (LC-NA) system in response to discrete stimuli or cognitive demands [42] [45]. In memory research, baseline pupil size itself is a meaningful metric; a larger baseline diameter preceding stimulus onset predicts successful long-term memory formation, supporting the "readiness to remember" (R2R) framework [10].

Experimental Protocol:

• Baseline Period Definition: The baseline is typically defined as a 500-millisecond to 1-second window immediately preceding the stimulus onset or a "Go" signal for a response [25] [46] [10]. Using the period before the previous trial's response cue can help avoid motor artifact contamination [46].
• Calculation: The average pupil diameter within this baseline window is calculated for each trial.
• Correction: This average baseline value is subtracted from every data point in the subsequent trial time series, resulting in a baseline-corrected pupil diameter trace expressed as a change from baseline (Δ pupil diameter).

Normalization

Normalization procedures control for inter-individual differences in baseline pupil size and anatomical constraints, reducing participant-to-participant variability and facilitating group-level analyses.

Physiological Rationale: Absolute pupil size varies substantially between individuals due to factors like age, iris pigmentation, and anatomy [47]. Since cognitive research is primarily concerned with relative changes in pupil size driven by the LC-NA system, normalization allows comparisons of the magnitude of cognitive responses across participants [46].

Experimental Protocols:

• Z-score Transformation: This is a common normalization method where the pupil diameter for each sample is converted to a z-score based on the distribution of all pupil data for that participant. The formula is ( z = (x - μ) / σ ), where ( x ) is the raw pupil diameter, ( μ ) is the mean diameter across the entire experiment for the participant, and ( σ ) is the standard deviation [46]. This expresses pupil size in terms of standard deviations from the participant's mean.
• Percent Signal Change: Alternatively, pupil diameter can be expressed as a percentage change from the baseline value for each trial. This is calculated as ( [(Diameter - Baseline) / Baseline] * 100 ).

Table 1: Comparison of Normalization Techniques in Pupillometry

Normalization Method	Formula	Application Context	Key Advantage
Z-score Transformation	( z = (x - μ) / σ )	Within-participant designs; comparing responses across different task conditions [46].	Removes inter-individual differences and stabilizes variance for group-level statistics.
Percent Signal Change	( \frac{(D - B)}{B} \times 100 )	Single-trial analyses; when the relative magnitude of change is of primary interest.	Intuitively represents the proportional magnitude of the pupillary response.

Statistical Modeling

Advanced statistical models are required to handle the high-dimensional, time-series nature of pupillometry data and to test hypotheses about cognitive processes in memory.

Dimensionality Reduction: Temporal Principal Components Analysis (tPCA) is used to identify latent, underlying physiological processes that shape the pupil signal over time. This technique decomposes the pupillary time series into a set of principal components (a "pupillary manifold"), which represent distinct temporal patterns of constriction and dilation [45]. The resulting component scores are manageable for statistical testing and are more physiologically interpretable than single time-point analyses.

Mixed-Effects Models: These models are the gold standard for analyzing pupillometry data because they can account for both fixed effects (experimental conditions of interest, like Memory_Outcome: Recalled vs. Forgotten) and random effects (variance due to individual participants and specific stimuli) [25] [10]. This is crucial for generalizing findings beyond the specific items and participants in a study.

Bayesian Frameworks: Bayesian statistics are increasingly employed, providing robust evidence for either the alternative or null hypothesis. For example, a Bayes Factor (BF₁₀) greater than 10 can constitute strong evidence that pupil size is larger during the encoding of later-recalled items compared to forgotten ones [10].

Experimental Protocols for Memory Research

Protocol: Production Effect and Memory

This protocol examines how vocal production enhances memory distinctiveness, using pupillometry to track the attentional processes involved [25].

• Hypothesis: Words read aloud will show better recall than words read silently, correlated with greater pupil dilation during encoding.
• Task Design:
  • Participants: Study a list of words under different instructions: Aloud (read word aloud), Silent (read word silently), and Control (say "check").
  • Design: Instructions can be presented concurrently with or preceding the word. A "Go" signal can be used to separate preparatory processing from the motor act of speaking [25].
  • Memory Test: A subsequent surprise recognition or free recall test.
• Pupillometry Analysis:
  • Compare baseline-corrected pupil diameter traces between Aloud and Silent conditions.
  • Correlate the magnitude of the "pupillometric production effect" (difference in dilation between Aloud and Silent) with the behavioral production effect (difference in memory accuracy) [25].

Protocol: Predicting Long-Term Memory from Pre-stimulus Pupil Size

This protocol tests the "readiness to remember" (R2R) hypothesis by investigating if pre-stimulus arousal predicts memory encoding success [10].

• Hypothesis: Successfully recalled words will be associated with larger baseline pupil size (500 ms before word onset) and larger pupil diameter during encoding than forgotten words.
• Task Design:
  • Participants study words for a subsequent memory test (free recall in Experiment 1; old/new recognition in Experiment 2) [10].
  • Pupil size is recorded throughout the study phase.
• Statistical Analysis:
  • Item-by-item ANOVA: Confirm that pupil size differences are not driven by specific word properties [10].
  • Bayesian t-tests: Compare mean baseline pupil size and encoding-phase pupil size between subsequently Recalled and Forgotten trials.
  • Moderation Analysis: Use a tool like the PROCESS macro for SPSS to test if factors like serial position moderate the link between baseline pupil size and memory [10].

Table 2: Key Dependent Variables and Analytical Tests for Memory Protocols

Protocol	Key Dependent Variables	Primary Statistical Analysis	Expected Outcome (Based on Cited Research)
Production Effect [25]	- Pupil diameter during preparation/encoding- Behavioral memory accuracy	- Mixed-effects models- Correlation analysis	Greater pupil dilation for Aloud vs. Silent words, correlating with memory benefit. BF₁₀ > 10 [10].
Readiness to Remember [10]	- Baseline pupil size (pre-stimulus)- Pupil size during encoding- Memory accuracy (Recall/Hit vs. Miss)	- Bayesian t-tests- Repeated-measures ANOVA- Moderation analysis	Larger baseline and encoding pupil size for recalled words. Effect sizes: Cohen's d ~ 0.60-0.95 for encoding period [10].

Signaling Pathways and Workflows

Neurobiological Signaling Pathway

The following diagram illustrates the primary neurobiological pathway linking cognitive tasks to changes in pupil size, which serves as an indirect measure of locus coeruleus activity in memory research.

Data Preprocessing and Analysis Workflow

A standardized preprocessing pipeline is essential for transforming raw pupil data into analyzable metrics. The following workflow synthesizes best practices from current research.

The Scientist's Toolkit: Research Reagent Solutions

Table 3: Essential Materials and Software for Pupillometry Research

Item / Solution	Function / Application	Example Products / Protocols
Research-Grade Eye Tracker	High-fidelity recording of pupil diameter and gaze position at high sampling rates (e.g., 1000 Hz).	EyeLink 1000 (SR Research) [46], Smart Eye Aurora [48].
Portable Pupillometry Device	Enables rapid, clinically-applicable protocols for specific reflexes like the Pupillary Light Reflex (PLR).	RETeval system [47].
Pupillometry Preprocessing Suite	Standardized, FAIR-principle-based software for robust and reproducible data preprocessing.	eyeris (https://github.com/shawntz/eyeris) [42].
Dimensionality Reduction Library	Identifies latent physiological processes from pupil time-series data via temporal PCA.	Pupilla R library (https://eblini.github.io/Pupilla/) [45].
Artifact Removal & Interpolation Algorithm	Corrects for blinks and missing data; cubic hermite interpolation preserves the natural shape of the pupil response [46] [48].	Median Absolute Deviation (MAD) for outlier detection; PCHIP interpolation [48].
Multimodal Data Integration Platform	Synchronizes pupillometry with other physiological and behavioral data streams (e.g., facial expression, EEG).	iMotions Platform [48].

Overcoming Methodological Pitfalls and Optimizing Pupillometry Signal Quality

The Critical Role of Baseline Period Selection and Duration

In pupillometry research within memory studies, the baseline period is not merely a technical control; it is a rich source of data that captures an individual's preparatory cognitive state. The Readiness to Remember (R2R) framework suggests that a person's ability to encode and retrieve information fluctuates from moment to moment, influenced by their initial state of attention, which is mediated by arousal levels [5]. Pupil size, driven by autonomic nervous system activity and closely linked to the firing of the Locus Coeruleus (LC), serves as a non-invasive proxy for this state. The baseline period, therefore, offers a window into this neurophysiological "readiness" before a stimulus is even presented. Consequently, the selection and duration of this period are critical methodological decisions that directly impact the validity and interpretability of research findings on attention and memory.

Theoretical Foundations and Key Evidence

The critical importance of the baseline period is robustly supported by empirical findings. Research demonstrates that baseline pupil size, measured before stimulus onset, can significantly predict subsequent long-term memory performance.

Evidence for Baseline Pupil Size Predicting Memory

A 2025 study specifically investigated whether pupil diameter before and during stimulus presentation predicts memory performance [5]. In two experiments involving a long-term memory test, the results consistently showed that larger pupil size during the baseline period (500 ms before word presentation) was associated with successfully recalled words in both free-recall and recognition tasks [5]. This provides strong evidence that the pre-stimulus physiological state, indicative of arousal and attentional readiness, influences encoding strength.

The Neurophysiological Basis

The phenomenon is grounded in the LC-NE system's activity:

• Tonic (baseline) pupil size is correlated with the baseline firing rate of LC neurons [5].
• The Adaptive Gain Theory posits an inverted U-shaped relationship between tonic LC activity (arousal) and task performance, with an optimal intermediate level for peak performance [5].
• This pre-stimulus arousal state, reflected in baseline pupil size, aligns with the R2R framework, positing that momentary attentional readiness affects recollection ability [5].

Table 1: Key Empirical Findings Linking Baseline Pupil Size to Memory Performance

Experiment	Memory Task	Baseline Finding	Statistical Evidence
Experiment 1 [5]	Free Recall	Larger baseline pupil size for recalled words	BF₁₀ = 72.18, Cohen’s d = 0.60
Experiment 2 [5]	Recognition	Larger baseline pupil size for "Hit" trials	BF₁₀ = 49.96, Cohen’s d = 0.70

Experimental Protocols and Application Notes

Properly designing and executing a pupillometry experiment requires meticulous attention to the baseline period. The following protocols provide a guideline for ensuring data quality and interpretability.

Protocol: Designing the Baseline Period

This protocol outlines the key considerations for establishing a baseline period in a trial-based pupillometry experiment, adapted from methodological guidelines [13].

• 1. Timing and Duration:

  • Placement: The baseline period must immediately precede the cognitive event of interest (e.g., stimulus presentation) [13].
  • Duration: A typical baseline window ranges from 100 ms to 500 ms before event onset [5]. The duration must be justified based on the research question and the need to capture a stable tonic state.

• 2. Environmental and Stimulus Control:

  • Visual Field: Maintain a uniform, constant visual display during the baseline period. The screen should consist of a neutral background (e.g., uniform mid-level gray) with only a central fixation point [13] [49].
  • Luminance: Ensure the luminance of the pre-trial screen is identical to the inter-stimulus interval and consistent across all trials and conditions. Any change in luminance between the baseline and the stimulus period will evoke a light reflex, confounding cognitive pupillary responses [13].

• 3. Participant Instruction:

  • Instruct participants to maintain fixation on a central point throughout the baseline and stimulus presentation periods to minimize eye movements and blinks that introduce artifact [49].
  • Encourage participants to blink during designated intervals (e.g., response periods) to minimize data loss during critical baseline and task periods.

Protocol: Preprocessing and Data Analysis

Once data is collected, a rigorous preprocessing pipeline is essential for isolating the task-evoked pupil response from the baseline signal.

• 1. Data Preprocessing:

  • Artifact Removal: Identify and remove periods contaminated by blinks, large saccades, or signal loss. Use linear interpolation to fill in small, missing data segments [13].
  • Filtering: Apply a low-pass filter (e.g., a median filter) to smooth high-frequency noise while preserving the slower, task-evoked pupillary response [13].

• 2. Baseline Correction:

  • For each trial, calculate the mean pupil size during the baseline period.
  • Subtract this baseline mean from all pupil-size values in the subsequent trial epoch (e.g., the stimulus presentation period). This yields a baseline-corrected pupil size, expressed as a change from baseline (Δ pupil size) [13].
  • This step helps control for slow drifts in tonic pupil size and isolates the phasic response to the trial event.

• 3. Statistical Analysis:

  • Use repeated-measures analyses (e.g., ANOVA) to compare baseline-corrected pupil responses between conditions (e.g., recalled vs. forgotten) [5].
  • To analyze the raw baseline size itself (as a measure of tonic arousal), employ trial-level analyses (e.g., mixed-effects models) to test if pre-stimulus pupil diameter predicts subsequent behavioral outcomes like memory accuracy [5].

Visualization of Workflows and Pathways

To elucidate the concepts and protocols discussed, the following diagrams provide a visual summary.

Diagram 1: Experimental workflow for baseline pupillometry in memory research.

Diagram 2: The neuro-cognitive pathway from baseline state to memory.

The Scientist's Toolkit: Research Reagent Solutions

Table 2: Essential Materials and Tools for Cognitive Pupillometry

Item/Tool	Function/Description	Example/Note
Eye Tracker	Measures pupil diameter and gaze position with high temporal resolution.	Systems from vendors like SR Research (EyeLink), Tobii, or SMI. Critical for capturing task-evoked responses [13].
Stimulus Presentation Software	Precisely controls the timing and content of visual/auditory stimuli.	MATLAB with Psychtoolbox, Python with PsychoPy, or E-Prime. Ensures millisecond precision [49].
Dedicated Display System	Presents visual stimuli under controlled luminance.	A linearized (gamma-corrected) monitor. For dichoptic presentation, a stereoscope may be used [49].
Data Preprocessing Toolbox	Scripts for artifact removal, filtering, and baseline correction of raw pupil data.	Custom Python or MATLAB scripts, such as those provided by [13]. Essential for standardizing analysis.
Statistical Analysis Software	Conducts trial-level and group-level analyses (e.g., ANOVA, mixed models).	R, Python (with Pandas, Statsmodels), SPSS, or JASP. Allows for modeling of baseline pupil size as a predictor [5].
Controlled Testing Environment	A dark, quiet room to minimize external influences on pupil size.	Essential for isolating cognitive, rather than sensory, pupillary responses [49].

Managing and Correcting for Common Data Artifacts

In pupillometry research investigating the link between attention and long-term memory, data quality is paramount. The pupil is a gateway to understanding cognitive processes such as the "readiness to remember" (R2R), a framework suggesting that a person's attentional state before encoding begins partially explains their ability to recollect information [5]. Research has demonstrated that larger baseline pupil size, even before a word is presented, predicts successful later retrieval, supporting the role of arousal and preparatory attention in memory formation [5]. However, the pupil's size is influenced by numerous factors beyond cognition, including blinks, head movements, and the pupillary light reflex. These factors introduce artifacts that can obscure genuine cognitive signals. Therefore, robust management and correction of data artifacts are not merely procedural steps but are critical to accurately interpreting the physiological correlates of attention and memory. This document outlines standardized protocols for identifying, preprocessing, and correcting common artifacts in pupillometry data within the context of memory research.

A systematic approach to artifact management begins with understanding their frequency and establishing clear, quantitative criteria for trial exclusion. The following table summarizes common artifacts and the thresholds used in contemporary research.

Table 1: Common Data Artifacts and Typical Exclusion Criteria

Artifact Type	Description	Common Detection Methods	Typical Exclusion Criteria
Eye Blinks	Temporary eyelid closure causing data loss or unrealistic pupil constriction.	Detection of rapid, large changes in pupil size or periods of missing data [43] [50].	Trials with >15% to >30% missing data from blinks are often excluded [50].
Rapid Dilation/Constriction	Physiologically implausible, sudden changes in pupil diameter, often from tracking errors.	Dilation speed outliers; absolute change between samples divided by their temporal separation [43].	Samples exceeding a normalized speed threshold (e.g., 3-4 standard deviations from the mean) are marked invalid [43].
Temporally Isolated Samples	Brief, valid-looking data points surrounded by long periods of missing data, likely noise.	Identification of single valid samples surrounded by missing data for a specified duration (e.g., >200 ms) [43].	Isolated samples are removed and treated as missing data.
Physiological Implausibility	Pupil size values outside the possible biological range for humans.	Feasibility check against predefined minimum and maximum diameters [43].	Values below 1.5 mm or above 9 mm for diameter are typically rejected [43].
Trend-Line Deviation	Samples that deviate significantly from the overall trend of the pupil waveform.	Comparison of individual samples to a moving average or smoothed trend line [43].	Samples exceeding a deviation threshold (e.g., 3-4 standard deviations) are marked invalid.

Experimental Protocols for Artifact Management

Protocol 1: A Standardized Preprocessing Pipeline

This protocol provides a step-by-step workflow for preprocessing pupil-size data prior to statistical analysis, based on established guidelines [43] [51].

1. Preparation of Raw Data:

• Objective: Convert eyetracker output into a standardized format for processing.
• Procedure: Import raw data, including pupil size time series for the left and/or right eyes and segmentation information (e.g., trial onset markers). Remove clearly invalid samples, such as those with non-positive values or those explicitly marked as invalid by the eyetracker [43].

2. Filtering and Artifact Detection:

• Objective: Identify and remove samples contaminated by noise.
• Procedure: Apply a sequential filter pipeline to the raw data:
  • Dilation Speed Outliers: Calculate the normalized dilation speed between consecutive samples. Flag samples where this speed exceeds a threshold of 3-4 standard deviations from the mean [43].
  • Trend-Line Deviation: Smooth the data (e.g., with a moving average) and flag samples that deviate significantly from this trend (e.g., >3 SD) [43].
  • Temporally Isolated Samples: Identify and remove single valid samples surrounded by missing data.
  • Feasibility Range: Reject samples outside the 1.5 mm to 9 mm range [43].

3. Up-Sampling and Smoothing:

• Objective: Increase temporal resolution and reduce high-frequency noise.
• Procedure: Up-sample the data to 1000 Hz to ensure one sample per millisecond, which simplifies subsequent timeseries analysis [51]. Smooth the data using a moving-window average (e.g., with a window of 51 ms) or a Hanning filter to attenuate high-frequency noise without distorting the underlying task-evoked response [51].

4. Interpolation:

• Objective: Reconstruct the pupil waveform over periods of missing data.
• Procedure: Use linear or cubic-spline interpolation to fill gaps created by removed artifacts. Linear interpolation is safer and adequate for calculating aggregate measures like mean dilation. Cubic-spline interpolation may recover a more realistic waveform but can sometimes produce unrealistic shapes. It is recommended to set a maximum gap length (e.g., 750 ms) over which interpolation is performed to avoid creating data from extensive periods of missing information [51].

5. Baseline Correction:

• Objective: Account for random fluctuations in tonic pupil size and improve statistical power for detecting task-evoked changes [52].
• Procedure: For each trial, calculate the median pupil size during a baseline period (e.g., 200-500 ms before stimulus onset). Apply a subtractive baseline correction (corrected pupil size = pupil size − baseline) [52] [51]. Subtractive correction is strongly recommended over divisive correction, as it is less distorted by unrealistically small baseline values caused by residual artifacts [52].

6. Evaluation of Missing Data:

• Objective: Ensure data quality on a per-trial basis before analysis.
• Procedure: Calculate the percentage of missing data for each trial after preprocessing. Establish an exclusion criterion (e.g., reject trials with >15% missing data) to remove trials that are too compromised for reliable analysis [51] [50].

Protocol 2: Experimental Design to Minimize Artifacts

Proactive experimental design can significantly reduce the occurrence of artifacts.

• Control Stimulus Luminance: As light is the primary driver of pupil size, ensure that the average brightness (luminance) of visual stimuli is perfectly matched across all experimental conditions. This is a fundamental requirement to prevent confounds in cognitive pupillometry [13].
• Minimize Head Movement: Use a chin rest or forehead bar to stabilize the participant's head position relative to the eyetracker.
• Instruct Participants: Before the experiment, instruct participants to blink normally but to try to avoid blinking during critical trial periods (e.g., stimulus presentation). Encourage them to sit as still as possible.
• Pilot Testing: Conduct pilot tests to ensure the eyetracker provides a stable signal and that the experimental paradigm is comfortable for participants, minimizing unnecessary movement and frustration.

Workflow Visualization

The following diagram illustrates the logical sequence of the preprocessing pipeline described in Protocol 1.

The Scientist's Toolkit: Essential Materials and Reagents

Table 2: Key Research Reagent Solutions and Essential Materials

Item	Function/Description	Application in Memory Research Context
Video-Based Eyetracker	A camera-based system (e.g., from SR Research, Tobii) that records pupil size at a high sampling rate (e.g., 250-1000 Hz).	Captures the raw pupil diameter time series during encoding and retrieval phases of memory tasks [5] [13].
Pupillometry Analysis Software	Custom scripts (Python, R, MATLAB) or specialized toolboxes (e.g., PupilMetrics) for automated data preprocessing and analysis [53] [13] [51].	Standardizes artifact removal and feature extraction, enabling comparison of baseline and task-evoked pupil responses between remembered and forgotten trials [5] [53].
Stimulus Presentation Software	Software (e.g., PsychoPy, E-Prime) to present auditory or visual stimuli and record behavioral responses with precise timing.	Presents memory task items (e.g., words, images) and collects recall/recognition accuracy data [5] [25].
Chin/Forehead Rest	A mechanical stabilizer to minimize head movement.	Reduces motion artifacts, ensuring a stable pupil signal, which is critical for measuring subtle baseline differences predictive of memory [5].
Dedicated Analysis Computer	A computer with sufficient processing power and memory (RAM) to handle large timeseries datasets.	Runs preprocessing pipelines and statistical models (e.g., multilevel regression) on high-resolution pupil data [13] [43].

Effective management of data artifacts is the foundation of reliable pupillometry research in attention and memory. By implementing the standardized protocols and preprocessing workflows outlined here, researchers can mitigate the influence of blinks, movements, and other sources of noise. This, in turn, ensures that the observed pupillary signals—such as the "readiness to remember" indicated by baseline dilation—are genuine reflections of underlying cognitive processes rather than mere measurement error. Adopting these rigorous practices is essential for advancing our understanding of the intricate relationship between attention, arousal, and long-term memory.

Addressing the Challenges of Anticipatory and Pre-stimulus Cognitive Processes

The investigation of anticipatory and pre-stimulus cognitive processes represents a paradigm shift in memory research, moving beyond the traditional focus on neural activity elicited by an event to encompass the critical preparatory processes that occur before an event is encountered. A growing body of evidence demonstrates that the brain's state preceding stimulus presentation significantly influences subsequent memory formation [54] [5]. This article frames these developments within the context of pupillometry measures of attention, presenting the "readiness to remember" (R2R) framework as a theoretical foundation for understanding how anticipatory processes shape memory encoding [5].

Pupillometry provides a non-invasive window into cognitive and autonomic processes, with pupil size fluctuations reflecting activity of the locus coeruleus-norepinephrine (LC-NE) system, a key neuromodulatory system regulating arousal and attention [13] [55]. Recent research establishes that baseline pupil size, measured before stimulus onset, can predict subsequent memory performance, suggesting that preparatory cognitive states are physiologically embedded in moment-to-moment fluctuations of pupil-linked arousal [5]. This article details experimental protocols and analytical frameworks for leveraging pupillometry to investigate these anticipatory processes in memory research.

Theoretical Framework: The Readiness to Remember (R2R) Hypothesis

The Readiness to Remember (R2R) framework posits that a person's physiological state of attentional preparedness before encountering information partially determines their ability to subsequently recollect that information, mediated by the strength of encoding [5]. This preparatory state fluctuates from moment to moment according to attention levels and is mediated by the LC-NE system [5] [55].

According to this framework, the initial state of attention, reflected in tonic pupil size, affects the ability to recollect information. The theoretical underpinnings of R2R align with the Adaptive Gain Theory, which suggests that baseline (tonic) pupil size correlates with LC neuron firing rate and success in attentional tasks, including learning [5]. This relationship follows an inverted U-shaped curve where intermediate arousal levels optimize performance, while both low and high levels impair it [5].

Table 1: Key Theoretical Concepts in Anticipatory Cognitive Processes

Concept	Description	Physiological Correlate
Readiness to Remember (R2R)	Moment-to-moment physiological preparedness for effective encoding	Tonic pupil size before stimulus onset [5]
Pupil-Linked Arousal	Cognitive arousal state reflecting LC-NE system activity	Tonic and phasic pupil diameter changes [13]
Anticipatory Task-Set Control	Cognitive preparation for upcoming task demands	Frontally-distributed negative-going ERP components [54] [56]
Inverted U-Shaped Performance	Optimal performance at intermediate arousal levels	Baseline pupil size predictive of memory recall [5]

Quantitative Evidence for Pre-stimulus Predictors of Memory

Recent studies provide compelling quantitative evidence that physiological measures before stimulus onset can predict subsequent memory performance. The following table summarizes key findings from seminal studies in this domain.

Table 2: Quantitative Evidence for Pre-Stimulus Predictors of Memory Formation

Study Paradigm	Pre-stimulus Measure	Memory Outcome	Effect Size	Statistical Evidence
Pupillometry (Free Recall) [5]	Baseline pupil size at 500ms before word presentation	Recalled vs. forgotten words	Cohen's d = 0.60	BF₁₀ = 72.18 (robust evidence against null)
Pupillometry (Recognition) [5]	Baseline pupil size at 500ms before word presentation	Hit vs. miss trials	Cohen's d = 0.70	BF₁₀ = 49.96 (robust evidence against null)
EEG Study [54]	Negative-going frontally-distributed ERP modulation	Later recollection of words	Not reported	Observed for both auditory and visual words
Task-Switching EEG [56]	Widespread posterior positivity & right anterior negativity (~300ms pre-stimulus)	Effective task-set reconfiguration	Correlated with behavioral preparation measures	Protracted component distinct from P3 or CNV

The relationship between pupil size and memory performance demonstrates complex dynamics. Research by Micula et al. revealed a U-shaped relationship wherein pupil dilation initially corresponded with enhanced memory function up to an optimal threshold, beyond which further dilation was associated with diminished recall abilities [5]. This pattern theoretically emerges from the allocation of cognitive resources, where excessive attentional resources devoted to stimulus identification may reduce the capacity available for memory encoding [5].

Experimental Protocols for Pupillometry in Memory Research

Protocol 1: Assessing Pre-stimulus Pupil Size and Memory Encoding

This protocol measures how baseline pupil size before stimulus presentation predicts subsequent memory performance.

Materials and Equipment:

• Eye tracker with pupil measurement capability (minimum 60Hz sampling rate)
• Stimulus presentation software (e.g., PsychoPy, E-Prime)
• Controlled lighting environment (<3 lux for consistent pupil baseline)
• Chin/forehead rest to minimize head movements

Procedure:

• Participant Preparation: Calibrate eye tracker following manufacturer specifications. Ensure ambient lighting remains constant throughout testing.
• Trial Structure:
  • Fixation cross (1000ms)
  • Baseline period (500ms) with blank screen
  • Stimulus presentation (word or image, 2000ms)
  • Retention interval (1000ms)
  • Distractor task (optional)
• Testing Session: Present 100-200 trials across multiple blocks with breaks.
• Memory Test (45-minute delay): Administer either:
  • Free recall: Participants write down all remembered items (5 minutes)
  • Recognition test: Present old and new items; participants classify as "old" or "new"

Data Analysis:

• Extract mean pupil diameter during the 500ms baseline period preceding each stimulus
• Segment trials into subsequently remembered vs. forgotten based on memory test
• Conduct paired-samples t-test or Bayesian equivalent to compare baseline pupil size between conditions
• Calculate effect sizes (Cohen's d) for magnitude of differences

Protocol 2: Chromatic Pupillometry for Sensory Contribution Assessment

This protocol adapts clinical chromatic pupillometry methods [57] [58] for basic memory research to dissect different photoreceptor contributions to preparatory states.

Materials and Equipment:

• Pupillometer capable of presenting calibrated chromatic stimuli
• Blue (469nm) and red (640nm) LED light sources
• Neutral density filters for intensity control
• Data processing software (e.g., Python with NumPy, SciPy)

Procedure:

• Light Calibration: Measure and match photon flux for blue and red stimuli.
• Stimulus Sequence:
  • 10s darkness (baseline recording)
  • 9s exponentially increasing blue light
  • 22s darkness (recovery period)
  • 9s exponentially increasing red light
  • 10s darkness (recovery period)
• Data Collection: Test both eyes separately; randomize starting eye.

Analysis Pipeline:

• Preprocess pupil data: remove blinks, interpolate missing values
• Calculate percentage change from baseline pupil size
• Extract pupillometric features: constriction latency, maximum constriction, recovery velocity
• Compare parameters between blue and red light conditions

Experimental Workflow for Pre-stimulus Pupillometry

The Scientist's Toolkit: Essential Research Reagents and Materials

Table 3: Essential Materials for Anticipatory Process Research

Item	Specifications	Research Function
Eye Tracking System	Minimum 60Hz sampling rate, infrared illumination	Precise measurement of pupil diameter with minimal interference [13]
Chromatic Stimulation Device	Blue (469nm) and red (640nm) LEDs with calibrated intensities	Selective targeting of photoreceptor types (rods/cones vs. melanopsin) [57] [58]
Stimulus Presentation Software	PsychoPy, E-Prime, or equivalent with millisecond precision	Accurate timing control for pre-stimulus intervals and stimulus presentation [13]
Data Processing Toolkit	Python with NumPy, SciPy; MATLAB with specialized toolboxes	Preprocessing of pupil data, artifact removal, and feature extraction [13]
Environmental Control System	Light meters (<3 lux control), sound-attenuated booth	Standardized testing conditions to minimize confounding variables [13]

Methodological Considerations and Best Practices

Experimental Design Considerations

When designing experiments investigating anticipatory processes, several critical factors must be addressed:

Luminance Control: As light is the main determinant of pupil size, brightness should be constant between conditions unless stimulus brightness is itself a manipulation of interest [13]. Even when brightness is controlled, other low-level differences between stimuli may affect pupil size, including visual change or movement (induces constriction), color differences, and distribution of brightness across the visual field [13].

Baseline Correction: Proper baseline correction accounts for differences in baseline pupil size between participants and across groups [58]. Express pupil size as percentage change from baseline rather than absolute measurements to control for individual differences.

Trial Structure: The cue-stimulus interval is critical for examining anticipatory processes. Research indicates that a minimum of 1.5 seconds is necessary to observe pre-stimulus effects on memory encoding [54].

Data Quality and Analysis Considerations

Artifact Handling: Implement semiautomated algorithms for blink artifact removal and data quality assessment [58]. Approximately 10-15% of trials may require exclusion due to blinks or other artifacts.

Multiple Comparisons: Given the time-series nature of pupillometry data, appropriate statistical corrections for multiple comparisons are essential. Cluster-based permutation tests or similar approaches are recommended [13].

Individual Differences: Account for factors known to affect pupil size, including age, medication use, and iris pigmentation, through careful screening and statistical control [55].

Neurophysiological Pathways of Pre-stimulus Processing

Clinical and Translational Applications

The investigation of anticipatory processes has significant implications beyond basic memory research. In clinical neurology and psychiatry, abnormalities in preparatory cognitive control may underlie memory deficits in conditions such as Alzheimer's disease, attention deficit disorders, and schizophrenia.

Chromatic pupillometry has demonstrated particular utility in clinical assessment. Recent studies show that handheld chromatic pupillometry (HCP) can accurately detect functional loss in glaucoma with an area under the ROC curve (AUC) of 0.94, sensitivity of 87.9%, and specificity of 88.4% [58]. The classification performance of HCP in early-moderate glaucoma (AUC=0.91) was similar to Humphrey Visual Field testing (AUC=0.91) and slightly reduced compared with OCT (AUC=0.97) [58].

In drug development, pupillometry measures of anticipatory processes offer objective biomarkers for evaluating cognitive-enhancing compounds. The R2R framework provides a theoretical basis for testing whether pharmacological agents can optimize preparatory states for memory encoding, potentially leading to novel therapeutic approaches for cognitive disorders.

The investigation of anticipatory and pre-stimulus cognitive processes represents a fundamental advancement in understanding memory formation. Pupillometry provides a powerful, non-invasive methodology for quantifying these preparatory states through the "readiness to remember" framework. The experimental protocols and analytical frameworks presented here offer researchers comprehensive tools for exploring how the brain's state before information encounter shapes subsequent memory. As research in this area progresses, the integration of pupillometry with other neurophysiological measures promises to further elucidate the complex temporal dynamics of memory encoding, with significant implications for both basic cognitive neuroscience and clinical applications.

Strategies for Participant Engagement to Prevent Fatigue-Related Signal Deterioration

In pupillometry research, fatigue effects present a significant challenge to data fidelity. This phenomenon is characterized by a substantial diminishment of the task-evoked pupil response over the course of an experiment, even when core task demands remain constant [59]. This attenuation of the phasic pupil response, which can exceed 60% over a single session, poses a direct threat to the validity of studies examining attention and memory encoding [59]. The physiological basis for this effect is linked to the locus coeruleus-norepinephrine (LC-NE) system. Tonic (baseline) and phasic (stimulus-evoked) activity within this system interact such that when overall arousal or task engagement wanes due to fatigue, the responsiveness of the pupil to discrete stimuli is correspondingly reduced [59]. For memory research utilizing pupillometry, this signal deterioration can directly confound the measurement of cognitive load and attentional engagement during encoding, potentially obscuring the very neural correlates of the "readiness to remember" (R2R) that researchers seek to capture [5].

The table below synthesizes key quantitative findings on the extent of pupillary fatigue and the efficacy of various engagement strategies.

Table 1: Quantified Fatigue Effects and Intervention Efficacy

Study Element	Key Quantitative Finding	Implication for Experimental Design
Overall Fatigue Effect	Task-evoked pupil response diminishes by at least 60% over an experiment's duration [59].	A strong, predictable fatigue effect must be accounted for in all statistical models.
Baseline Pupil & Memory	Larger baseline pupil size (before stimulus) predicts subsequent recall (Cohen's d = 0.60-0.70) [5].	Baseline pupil size is a meaningful metric of pre-stimulus cognitive state ("readiness to remember").
Social Break Efficacy	Breaks with social interaction were the only intervention to significantly reduce the rate of pupillary fatigue across trials [59].	Social engagement is a particularly potent method for sustaining participant arousal and task engagement.
Researcher Observation	Participants who were knowingly observed had larger overall peak pupil responses [59].	Experimenter presence (the "good subject effect") can systematically influence physiological measures.

Detailed Experimental Protocols for Engagement

This section provides detailed methodologies for key experiments investigating engagement strategies.

Protocol: Evaluating Break Interventions on Pupillary Fatigue

This protocol is designed to test the effect of different break types on mitigating the fatigue of the pupil response [59].

• Objective: To determine whether breaks, and specifically the social quality of breaks, can reduce the attenuation of the task-evoked pupil response over time.
• Participants: Assign participants to one of three between-subject conditions.
• Stimuli & Task: A cognitively demanding task (e.g., an n-back task, a paired-associate memory encoding task) with consistent difficulty throughout to isolate fatigue from condition effects [59].
• Conditions:
  • Control Group: Completes the entire task sequence without any breaks.
  • Kinetic Break Group: Receives scheduled breaks during which they engage in a non-social, tactile activity (e.g., playing with a simple, silent toy like a stress ball or thinking putty).
  • Social Break Group: Receives scheduled breaks of the same duration that involve casual, unstructured conversation with the research assistant.
• Pupillometry Measures:
  • Primary: Peak and mean task-evoked pupil diameter for each trial block.
  • Secondary: Baseline (tonic) pupil diameter at the start of each trial block.
• Analysis: Use linear mixed-effects models or growth curve analysis to model the change in phasic pupil response over time (trial block) as a function of the experimental group. The critical test is the interaction between time and group.

Protocol: Assessing the "Readiness to Remember" (R2R) via Baseline Pupil Size

This protocol examines how pre-stimulus physiological states, influenced by engagement level, predict long-term memory formation [5].

• Objective: To test the hypothesis that baseline pupil size, as an index of pre-trial arousal and attentional readiness, predicts subsequent memory performance.
• Participants: Psychology students or other adult cohorts.
• Procedure:
  • Study Phase: Participants are presented with a series of words (e.g., 30 items) on a screen. Each trial begins with a fixation cross.
  • Pupillometry Recording: Eye-tracking data is collected throughout, with a critical focus on the 500 ms window before word presentation (baseline) and the period during word presentation.
  • Memory Test: After a delay, memory is tested via either:
    • Experiment 1 (Free Recall): Participants verbally recall as many words as possible.
    • Experiment 2 (Recognition): Participants are shown "old" and "new" words and must classify them.
• Data Segmentation: Trials are sorted post-hoc into "recalled" (Hit) and "forgotten" (Miss) based on memory test performance.
• Analysis: Compare the average pupil size during the pre-stimulus baseline period and the stimulus presentation period between recalled and forgotten trials using Bayesian or standard frequentist statistics (e.g., ANOVA). A successful replication of the R2R effect would show significantly larger pupil size for recalled items in both time windows [5].

Visualizing Workflows and Physiological Pathways

Physiological Pathway from Fatigue to Signal Deterioration

The following diagram illustrates the proposed neurocognitive pathway through which fatigue impairs the pupillary response, a key measure in cognitive experiments.

Experimental Protocol for Engagement Strategies

This workflow outlines the procedural steps for evaluating different participant engagement strategies, such as social breaks, within a pupillometry study on memory.

The Scientist's Toolkit: Essential Research Reagents & Materials

Table 2: Key Materials and Reagents for Pupillometry-based Memory Research

Item	Function/Application in Research
Remote Eye-Tracker	Apparatus for capturing pupil diameter and gaze position with high temporal resolution without requiring a chin rest, allowing for more naturalistic testing environments.
Standardized Word Lists	Linguistic stimuli (e.g., nouns with matched frequency, concreteness, length) used in memory encoding tasks to control for the effect of a word's inherent properties on pupil response and recall [5].
Analysis Software (e.g., R, Python)	Platforms for implementing linear mixed-effects models and growth curve analysis to statistically account for and model the systematic change in pupil response over time [59].
Protocol for Social Interaction	A standardized script or set of guidelines for research assistants to follow during social breaks to ensure consistency in participant engagement across the study [59].
'Kinetic Break' Materials	Simple, non-electronic tactile toys (e.g., stress balls, putty) used as an active control during breaks to isolate the effect of social interaction from mere physical disengagement [59].

Pupillometry, the measurement of pupil size and reactivity, has emerged as a powerful, non-invasive tool to investigate cognitive processes such as attention, mental effort, and working memory [13]. Its application in memory research is particularly valuable as it provides a real-time, physiological index of cognitive load and resource allocation [60] [55]. However, the successful implementation of pupillometry requires carefully tailored protocols that account for the specific physiological and cognitive characteristics of different populations. This application note provides a detailed framework for optimizing pupillometry protocols for aging individuals and clinical groups within the context of memory research, ensuring data reliability and validity.

Population-Specific Considerations

Aging Populations

Cognitive aging brings about changes that directly impact pupillary responses. Research indicates that older adults exhibit different pupillary dynamics compared to younger adults during cognitive tasks, which must be considered in experimental design.

Key Findings and Protocol Adjustments for Aging Populations:

• Increased Cognitive Load Sensitivity: Older adults show a steeper increase in pupil dilation slope under high cognitive load in tasks like word-span recall, suggesting greater cognitive effort [60]. Protocols should include task difficulty levels that are appropriately challenging without causing floor or ceiling effects.
• Baseline Pupil Size: A trend for decreased baseline pupil size and maximum constriction with age has been observed, though correlations are not always statistically significant [47]. Age should be a covariate in analyses.
• Stimulus Selection: Word-span tasks, which impose greater linguistic and semantic demands, may be more sensitive for detecting age-related differences than digit-span tasks [60]. Consider using semantically complex stimuli for sensitive assessment of memory load.
• Post-Illumination Pupillary Response (PIPR): The PIPR, a sustained constriction driven by melanopsin-containing retinal ganglion cells, appears to be independent of age [47]. This makes it a stable measure for isolating cognitive from purely visual effects in aging studies.

Clinical Populations

Clinical groups, such as those with mild traumatic brain injury (mTBI) or neurodegenerative conditions, present unique challenges and opportunities for pupillometric assessment.

Key Findings and Protocol Adjustments for Clinical Groups:

• Subtle Neurological Dysfunction: In mTBI, even with normal Glasgow Coma Scale (GCS) scores, patients with subarachnoid hemorrhage show significantly lower average pupillary dilation velocity (0.7 mm/s vs. 1.1 mm/s) [12]. Quantitative pupillometry can detect these subtle impairments where traditional exams fail.
• Pupillary Velocity as a Biomarker: Strong positive correlations exist between pupillary constriction velocity and GCS scores (Sr = 0.9), making it a sensitive prognostic indicator [12]. Incorporate velocity metrics as primary outcomes in clinical studies.
• Differentiating Clinical Conditions: Amnestic Mild Cognitive Impairment (MCI) individuals demonstrate greater pupil dilation during digit-span tasks compared to cognitively normal older adults, suggesting pupillometry's potential as an early biomarker for neurodegenerative disease [60]. Use pupillometry to stratify risk within clinical groups.

Table 1: Key Pupillometry Findings and Recommendations for Different Populations

Population	Key Finding	Protocol Recommendation
Aging Adults	Higher pupil dilation slope under high load (word-span) [60]	Use word-span tasks; include multiple difficulty levels.
Aging Adults	Trend for smaller baseline & constriction size [47]	Control for age in analysis; ensure proper dark adaptation.
mTBI Patients	Reduced dilation velocity (0.7 vs 1.1 mm/s) with SAH [12]	Focus on pupillary velocity metrics, not just size.
Amnestic MCI	Greater dilation during memory tasks [60]	Use as a sensitive measure for early cognitive decline.

Detailed Experimental Protocols

Protocol A: Cognitive Load and Working Memory Assessment

This protocol is adapted from studies investigating aging and short-term memory, using tasks that systematically manipulate cognitive load [60].

Aim: To measure task-evoked pupil responses to increasing cognitive demands in working memory. Population: Older adults, clinical groups with cognitive deficits. Materials: Eye tracker, audio presentation system, behavioral response collection.

Procedure:

• Preparation & Calibration: Participant sits in a chin rest. Calibrate eye tracker. Ensure ambient lighting is consistent and dim.
• Dark Adaptation: A 3-minute dark adaptation period is sufficient for pupil stabilization in most cognitive studies [47].
• Task Structure:
  • Stimuli: Use digit-span and word-span tasks.
  • Trials: Each task comprises three levels of cognitive load (e.g., 3-, 5-, 7-digit spans; 3-, 4-, 5-word spans). Present 10 trials per level [60].
  • Trial Timeline:
    • Fixation: 1000 ms.
    • Encoding: Auditory presentation of digits/words. Onset of first item marks time zero.
    • Retention: Blank screen for 2000-3000 ms. Pupil size during this period is critical for aging studies [60].
    • Recall: Participant verbally recalls items in order.
• Data Recording: Record pupil diameter at 120 Hz or higher. Record behavioral accuracy.

Protocol B: Rapid Clinical Pupillometry Protocol

This protocol is designed for efficient, reliable assessment in clinical settings or with populations unable to tolerate long testing sessions [47].

Aim: To obtain robust measures of the pupillary light reflex (PLR), including transient and sustained components. Population: All, particularly suitable for clinical groups. Materials: Portable pupillometer (e.g., RETeval or NeurOptics NPi-200/300).

Procedure:

• Setup: Test one eye (e.g., right eye). Place device eyecup over the eye.
• Dark Adaptation: 3 minutes.
• Stimulation: Present a 500 ms, nearly full-field chromatic flash (e.g., 470 nm blue light). Stimulus retinal illuminance should be fixed in Trolands (e.g., 12,000 Td) to account for individual differences in baseline pupil size [47].
• Recording: The device's infrared camera records pupil size for several seconds (e.g., 8-10 s) post-stimulus.
• Repetition: Repeat 3-5 times per eye with inter-stimulus intervals of >30 seconds to avoid adaptation.
• Key Metrics: Extract Baseline size, Maximum Pupil Constriction, Constriction/Dilation Velocity, and Post-Illumination Pupillary Response.

Data Preprocessing and Analysis Workflow

Raw pupillometry data is noisy and requires robust preprocessing before analysis. The following workflow, synthesized from current guidelines, ensures data quality [48] [13].

Figure 1: Pupillometry data preprocessing workflow.

Step-by-Step Preprocessing Instructions:

• Data Cleaning:

  • Remove rows with null pupil diameter data and duplicate timestamps [48].
  • Filter out physiologically impossible values (e.g., diameters <2 mm or >8 mm) often caused by blinks or tracking loss [48].

• Interpolation:

  • Fill short, missing data segments (e.g., from blinks) using methods like Piecewise Cubic Hermite Interpolating Polynomial (PCHIP) to preserve the shape of the pupil response [48].

• Outlier Detection:

  • Identify unrealistic, rapid changes using a dilation-speed based method [48].
  • Calculate the normalized dilation speed metric: d′i = max( |di - di-1| / |ti - ti-1|, |di+1 - di| / |ti+1 - ti| ).
  • Apply a threshold based on the Median Absolute Deviation (MAD): Threshold = median(d′) + n * MAD, where n is a scaling factor (e.g., 3) [48].

• Filtering:

  • Apply a low-pass filter (e.g., moving average) to reduce high-frequency noise while preserving the slower task-evoked pupil response [60] [48].

• Trial Epoching & Baseline Correction:

  • Segment data into trials time-locked to event onsets (e.g., stimulus presentation).
  • Calculate the mean pupil size during a pre-stimulus baseline period (e.g., 200-500 ms before stimulus onset) and subtract this from each trial to isolate task-evoked changes [13].

The Scientist's Toolkit: Research Reagent Solutions

Table 2: Essential Materials and Equipment for Pupillometry Research

Item	Function/Description	Example Use Case
Research-Grade Eye Tracker	High-speed (≥120 Hz) camera with infrared illumination to record pupil size with high precision.	Measuring subtle, trial-evoked pupil responses during cognitive tasks [13].
Automated Pupillometer (e.g., NPi-200/300)	Hand-held, clinical-grade device providing objective, standardized PLR metrics like NPi, latency, and velocity.	Rapid assessment in clinical populations; excellent inter-rater reliability [12] [61].
Stimulus Presentation Software	Software (e.g., PsychoPy, E-Prime) for precise control of task timing and synchronization with the eye tracker.	Presenting digit-span or word-span tasks with millisecond accuracy [60].
Data Preprocessing Pipeline	Custom scripts (e.g., in Python, MATLAB) for artifact removal, interpolation, and filtering.	Implementing the standardized workflow in Figure 1 to ensure data quality [48] [13].
Chromatic Stimulation Device	Device capable of delivering full-field, wavelength-specific light flashes (e.g., 470 nm blue light).	Isecting the contributions of different retinal photoreceptors to the PLR [47].

Optimizing pupillometry protocols for aging and clinical populations is paramount for advancing memory research. Key considerations include selecting cognitively appropriate tasks, accounting for physiological differences in baseline pupillary function, and employing rigorous, standardized preprocessing methods. The protocols and guidelines detailed in this application note provide a foundation for generating reliable, interpretable, and valid pupillometric data, thereby strengthening its utility as a sensitive measure of attention and cognitive load in diverse populations.

Validating Pupillary Measures and Comparative Analysis with Other Biomarkers

Pupillometry, the measurement of pupil diameter, serves as a robust, non-invasive physiological index of cognitive processes fundamental to memory formation. Changes in pupil size are linked to activity of the locus coeruleus-norepinephrine (LC-NE) system, which regulates arousal and attention—critical components for successful encoding and retrieval. This protocol outlines methods for validating pupillary responses by correlating them with behavioral performance on memory tasks, thereby establishing pupillometry as a convergent measure within memory research frameworks. The "readiness to remember" (R2R) hypothesis suggests that cognitive states preceding stimulus presentation can influence encoding success, a premise that can be tested through pre-stimulus baseline pupillary measurements [5].

Theoretical Framework and Neurocognitive Pathways

Pupillary changes during cognitive tasks are governed by autonomic nervous system activity. During high-level arousal, the pupils dilate due to intense secretion of norepinephrine (NE) from the Locus Coeruleus (LC) via sympathetic pathways. Dopamine release in arousing states (e.g., motivation, reward, novelty) exerts excitatory action on wakefulness-promoting neurons of the LC, resulting in increased alertness. The Adaptive Gain Theory further suggests that baseline (tonic) pupil size correlates with LC neuron firing rate and success in attentional tasks, including learning. The relationship between arousal and performance follows an inverted U-shaped curve: optimal performance occurs at intermediate arousal levels, while both low (unfocused) and high (overstimulated) levels lead to performance decline [5].

Figure 1: Neurocognitive Pathway Linking Pupillary Response to Memory Performance. This diagram illustrates the proposed pathway through which a cognitive stimulus triggers locus coeruleus activity, leading to increased arousal, pupil dilation, and subsequent memory formation. The dashed line represents the correlational relationship validated in this protocol.

Experimental Protocols for Convergent Validation

Free Recall Memory Task (Experiment 1)

Objective: To examine the relationship between pupil diameter during encoding and subsequent performance on a free recall memory test.

Participants: Ninety-five psychology students participated in the original study. Sample size should be determined via power analysis for replication studies [5].

Stimuli and Apparatus:

• Stimuli: 30 words presented individually.
• Eye Tracking: Eye tracker to monitor pupil diameter at a minimum sampling rate of 60Hz (higher rates of 120-1000Hz are common in pupillometry research) [62].
• Software: Software for stimulus presentation (e.g., E-Prime, PsychoPy) that can send triggers to mark stimulus onset in the pupil data.

Procedure:

• Calibration: Perform a standard eye-tracker calibration and validation procedure.
• Baseline Period: Present a fixation cross for 1,500 ms. Record baseline pupil diameter during the 500 ms immediately preceding word onset.
• Encoding Phase: Present each word for a fixed duration (e.g., 3,000 ms). Record pupil diameter continuously throughout the trial.
• Distractor Task: Implement a brief (e.g., 2-minute) arithmetic task to prevent rehearsal.
• Free Recall Test: Provide participants with 5 minutes to verbally recall and write down as many words from the study phase as possible, in any order.

Data Preprocessing & Analysis:

• Pupil Data: Preprocess pupil data by applying a blink detection and interpolation algorithm. Filter the data (e.g., low-pass filter). Average pupil diameter for each trial across the baseline period and the entire stimulus presentation period.
• Trial Categorization: For each participant, categorize encoding trials based on subsequent memory performance: "Recalled" vs. "Forgotten."
• Statistical Analysis: Use a paired-samples t-test or Bayesian equivalent to compare average pupil size (separately for baseline and encoding periods) between "Recalled" and "Forgotten" trials. The original study found robust evidence for larger pupils in recalled trials (Baseline: BF₁₀ = 72.18, Cohen’s d = 0.60; Encoding: BF₁₀ = 86.58, Cohen’s d = 0.61) [5].

Recognition Memory Task (Experiment 2)

Objective: To replicate the pupillometry-memory relationship using a recognition test, assessing its generalizability across different memory retrieval processes.

Participants, Stimuli, and Apparatus: As described in Protocol 3.1.

Procedure:

• Calibration, Baseline, and Encoding Phase: Identical to Protocol 3.1.
• Recognition Test: Intermix the 30 "old" words with 30 new, unstudied "lure" words. Present words one at a time. For each word, participants indicate "Old" or "New" via button press.
• Data Categorization: Categorize encoding trials based on recognition test performance: "Hit" (studied word correctly identified as "Old") vs. "Miss" (studied word incorrectly identified as "New").

Data Preprocessing & Analysis: Analysis follows the same steps as Protocol 3.1, comparing "Hit" and "Miss" trials. The original study again found larger pupil size for remembered items (Baseline: BF₁₀ = 49.96, Cohen’s d = 0.70; Encoding: BF₁₀ = 1583.24, Cohen’s d = 0.95) [5].

Figure 2: Experimental Workflow for Convergent Validation. This flowchart outlines the two primary experimental protocols for correlating pupillary responses with behavioral memory performance, culminating in a shared data analysis step.

Table 1: Summary of Key Pupillometry Findings from Free Recall and Recognition Memory Experiments

Experimental Condition	Subsequent Memory Contrast	Baseline Pupil Size (500ms pre-stimulus)	Pupil Size During Encoding
Free Recall [5]	Recalled vs. Forgotten	BF₁₀ = 72.18, Cohen's d = 0.60	BF₁₀ = 86.58, Cohen's d = 0.61
Recognition [5]	Hit vs. Miss	BF₁₀ = 49.96, Cohen's d = 0.70	BF₁₀ = 1583.24, Cohen's d = 0.95
Mean Behavioral Accuracy
Free Recall	36.82% (SD = 14.12%)	Not Applicable	Not Applicable
Recognition (Hit Rate)	65.66% (SD = 12.47%)	Not Applicable	Not Applicable

Table 2: The Scientist's Toolkit - Essential Research Reagents and Materials

Item	Specification / Example	Primary Function in Protocol
Eye Tracker	Video-based, high temporal resolution (≥60 Hz, 120-1000 Hz ideal) [62]	Records continuous pupil diameter with sufficient precision to capture cognitive responses.
Stimulus Presentation Software	E-Prime, PsychoPy, OpenSesame	Presents experimental stimuli and sends synchronization triggers to the eye tracker.
Pupillometry Preprocessing Toolbox	Custom scripts (e.g., Python, R), Open-source pipelines (e.g., Open-DPSM [63])	Performs blink detection/interpolation, filtering, and baseline correction of raw pupil data.
Statistical Analysis Software	R, Python, JASP, SPSS	Conducts statistical comparisons (e.g., t-tests, Bayesian t-tests, mixed-effects models) between subsequent memory conditions.
Validated Word Lists	Normed for frequency, concreteness, emotional valence [5]	Provides standardized stimuli for memory tasks to control for confounding effects of word properties.

Methodological Considerations and the Multiverse Approach

Pupillometry data analysis involves multiple preprocessing steps where researcher decisions can influence results. A "multiverse analysis" approach—running analyses across a range of plausible preprocessing pipelines—is recommended to assess result robustness. Key decision points include [62]:

• Handling of Extreme Values: Thresholds for defining and removing artifacts.
• Blink Management: Interpolation method and window size.
• Baseline Correction: Time window selection for baseline period (e.g., -500 ms to 0 ms before stimulus).
• Participant Inclusion/Exclusion: Criteria based on data quality (e.g., percentage of valid pupil samples).
• Statistical Modeling: Incorporation of smoothers and random effects structure in mixed models to account for time-series data and individual variability.

Furthermore, environmental factors must be controlled. Luminance is a primary driver of pupil size; ensure consistent lighting and monitor brightness throughout the experiment. Stimuli should be matched for luminance where possible. The testing environment should be quiet and free from distractions to minimize stress, which can confound task-evoked pupillary responses [5].

These application notes provide a validated framework for using pupillometry as a convergent measure in memory research. The protocols demonstrate that pupil diameter, both during baseline and stimulus encoding, reliably predicts long-term memory performance. This relationship is robust across different testing modalities (free recall and recognition), strengthening the case for pupillometry as an objective, physiological index of the cognitive processes underlying memory formation. Adhering to these detailed protocols, while considering the methodological nuances of pupillometry data, will allow researchers to effectively leverage this tool in basic cognitive research and applied drug development contexts.

Pupillometry as a Sensitive Measure Where Behavioral Performance Reaches Ceiling

In cognitive research, many experimental paradigms face a fundamental limitation: behavioral performance measures (e.g., accuracy, reaction time) often ceiling effects when tasks are insufficiently challenging, obscuring meaningful cognitive processing differences. Pupillometry—the measurement of pupil diameter fluctuations—emerges as a powerful physiological correlate of cognitive processes that remains sensitive even when behavioral measures saturate. This sensitivity stems from pupil size's direct link to the locus coeruleus-norepinephrine (LC-NE) system, which regulates arousal, attention, and cognitive effort [13]. Within memory research, pupillometry provides a unique window into attentional allocation and encoding processes that predict subsequent memory performance, independent of overt behavioral responses [5].

The "readiness to remember" (R2R) framework posits that pre-stimulus brain states, mediated by attentional preparedness, significantly influence memory encoding success [5]. Recent evidence demonstrates that baseline pupil size before stimulus presentation—an indicator of tonic LC-NE activity—differentiates subsequently remembered from forgotten items, revealing preparatory cognitive states that behavioral measures cannot capture [5]. This application note details how pupillometry can be implemented in memory research to uncover these hidden cognitive dynamics when behavioral measures reach their ceiling.

Key Experimental Evidence & Quantitative Data

Pupillometric Predictors of Long-Term Memory Formation

Recent research establishes that pupil diameter during both baseline and stimulus encoding phases robustly predicts subsequent memory performance. The following table synthesizes key quantitative findings from seminal studies:

Table 1: Pupillometry Measures Predicting Subsequent Memory Performance

Experimental Paradigm	Participant Number	Pupillometric Measure	Memory Performance	Key Finding	Effect Size (Cohen's d)
Free Recall [5]	95 psychology students	Baseline pupil size (500ms pre-stimulus)	36.82% mean accuracy (SD=14.12%)	Larger baseline pupil for recalled vs. forgotten words	0.60
Free Recall [5]	95 psychology students	Pupil size during encoding	36.82% mean accuracy (SD=14.12%)	Larger encoding pupil for recalled vs. forgotten words	0.61
Recognition Memory [5]	95 psychology students	Baseline pupil size (500ms pre-stimulus)	65.66% mean hit rate (SD=12.47%)	Larger baseline pupil for "hit" vs. "miss" trials	0.70
Recognition Memory [5]	95 psychology students	Pupil size during encoding	65.66% mean hit rate (SD=12.47%)	Larger encoding pupil for "hit" vs. "miss" trials	0.95
Generation Effect [11]	4 experiments	Pupil size during generative encoding	Enhanced memory for self-generated information	Greater pupil dilation during generation vs. reading	N/A

Neural Mechanisms and Signal Interpretation

Pupil diameter fluctuations are governed by the autonomic nervous system, with cognitive effort-induced dilation primarily mediated by norepinephrine release from the locus coeruleus (LC) [5]. The relationship between pupil size and cognitive performance follows an inverted U-shaped curve, where optimal performance occurs at intermediate arousal levels [5]. This physiological basis ensures that pupillary responses provide a genuine index of cognitive resource allocation rather than stimulus characteristics, particularly when experiments carefully control for non-cognitive confounds like luminance [13].

Table 2: Interpretation of Pupillometric Responses in Cognitive Research

Pupillometric Measure	Physiological Correlate	Cognitive Interpretation	Research Application
Tonic (Baseline) Pupil Size	Tonic LC-NE activity, arousal state	"Readiness to encode," attentional preparedness	Predicting subsequent memory performance before stimulus onset
Phasic Task-Evoked Dilation	Phasic LC-NE response, cognitive resource allocation	Mental effort, task engagement, processing load	Measuring cognitive load differences despite equal behavioral performance
Peak Dilation Amplitude	Intensity of neuromodulatory response	Cognitive demand, surprise, novelty detection	Identifying thresholds of cognitive processing
Dilation Time Course	Temporal dynamics of LC-NE response	Processing stages, task difficulty dynamics	Differentiating automatic vs. controlled processing

Experimental Protocols & Methodologies

Core Protocol: Predicting Long-Term Memory Formation

This protocol outlines the methodology for using pupillometry to predict long-term memory formation, adapted from recent research demonstrating the "readiness to remember" effect [5].

Experimental Design

• Participants: 95 psychology students (or sample size with sufficient power for within-subjects design)
• Stimuli: 30 words matched for frequency, length, and concreteness
• Memory Task: Mixed design with free recall (Experiment 1) and recognition (Experiment 2) conditions
• Timing Parameters:
  • Baseline period: 500ms pre-stimulus fixation
  • Encoding phase: 3000ms word presentation
  • Retention interval: 15-30 seconds filled with distractor task
  • Recall/recognition test: 2-3 minutes duration

Procedure

• Calibration: Eye tracker calibration using 5-point grid
• Baseline Recording: 500ms fixation cross preceding each word
• Encoding Phase: Word presentation (3000ms) with continuous pupil recording
• Distractor Task: Arithmetic problems during retention interval
• Memory Test:
  • Free recall: 2 minutes to write all remembered words
  • Recognition: "Old/new" judgment for 60 words (30 old, 30 new)

Data Collection Parameters

• Sampling Rate: ≥ 100Hz (500Hz recommended)
• Pupil Metrics:
  • Baseline pupil diameter (mean of 500ms pre-stimulus)
  • Task-evoked pupil response (mean diameter during encoding)
  • Maximum dilation amplitude during encoding

Control Measures and Confound Management

Successful pupillometry requires rigorous control of non-cognitive influences on pupil size:

Table 3: Essential Control Measures for Cognitive Pupillometry

Confounding Factor	Impact on Pupil Size	Control Method
Luminance	Primary determinant: brighter light causes constriction	Use constant luminance across conditions; neutral background
Stimulus Color	Different wavelengths cause varying constriction	Use grayscale stimuli or balance colors across conditions
Visual Change/Movement	Induces pupil constriction	Minimize abrupt onsets/offsets; use smooth transitions
Blinking	Creates data gaps and measurement artifacts	Implement blink detection and interpolation algorithms
Respiratory Sinus Arrhythmia	Causes low-frequency pupil oscillations	Record in controlled environments; measure baseline periods

Visualization of Experimental Workflow and Neural Pathways

Neural Pathways of Cognitive Pupillometry

Experimental Workflow for Memory Prediction Studies

The Researcher's Toolkit: Essential Materials & Methods

Research Reagent Solutions

Table 4: Essential Materials and Software for Cognitive Pupillometry Research

Category	Specific Product/Software	Function/Purpose	Key Features
Eye Tracking Hardware	SR Research EyeLink, Tobii Pro, Pupil Labs	High-speed pupil imaging	500-1000Hz sampling, infrared illumination, high spatial resolution
Stimulus Presentation	PsychoPy, E-Prime, Presentation	Experimental control	Precise timing, synchronization with eye tracker, randomization
Data Preprocessing	PupilMetrics [40], Python (PyPillometry)	Artifact detection and removal	Blink interpolation, noise filtering, baseline correction
Statistical Analysis	R, Python (statsmodels), SPSS	Advanced statistical modeling	Mixed-effects models, Bayesian statistics, time-series analysis
Quality Control	NeurOptics PLR-200 [26]	Validation of pupillometer accuracy	0.05mm precision, standardized conditions, normative comparisons

Data Processing Pipeline

A standardized preprocessing workflow is essential for reliable pupillometry data:

• Raw Data Import: Import time-synchronized pupil diameter and experimental events
• Artifact Detection:
  • Identify blinks using velocity-based algorithms [40]
  • Flag signal drops using deviation thresholds
• Data Cleaning:
  • Remove blink-affected segments (typically ±150ms around blink)
  • Interpolate missing data using cubic spline or linear methods
• Filtering: Apply low-pass filter (typically 4-10Hz cutoff) to reduce high-frequency noise
• Baseline Correction: Subtract pre-stimulus baseline (mean of -500 to 0ms) from trial data
• Trial Averaging: Group trials by condition for statistical comparison

Application Notes for Specific Research Scenarios

Overcoming Ceiling Effects in Memory Research

When behavioral performance approaches ceiling (e.g., >90% accuracy), pupillometry reveals nuanced cognitive differences:

Implementation Strategy:

• Focus on pre-stimulus baseline pupil diameter as an indicator of encoding preparedness [5]
• Analyze pupil dilation timecourse rather than just peak amplitude
• Examine trial-by-trial covariation between pupil size and response time, even when accuracy is uniformly high

Case Example: In a word recognition paradigm where accuracy reached 94% across conditions, baseline pupil size successfully differentiated strong vs. weak memory traces, with larger pre-stimulus dilation predicting faster recognition latencies despite equivalent accuracy [5].

Specialized Applications

Phonological Contrast Detection: Pupillometry detects sensitivity to phonological contrasts even when behavioral discrimination fails, with pupil dilation reflecting higher-level linguistic processing rather than low-level acoustic perception [64].

Mental Effort Assessment: The generation effect (better memory for self-generated information) shows correlated pupil dilation increases, confirming mental effort allocation despite equivalent behavioral performance across conditions [11].

Perceptual Switching: During bistable perception tasks, pupil diameter peaks precisely at perceptual switch points, revealing neuromodulatory involvement in perceptual reorganization independent of behavioral reports [65].

Pupillometry provides cognitive researchers with a powerful, non-invasive tool for investigating memory and attention processes that remain invisible to conventional behavioral measures. By implementing the protocols and methodologies outlined in this application note, researchers can:

• Detect cognitive state differences when behavioral performance saturates
• Predict subsequent memory success from pre-stimulus physiological states
• Quantify mental effort allocation across experimental conditions
• Track temporal dynamics of cognitive processing with high temporal resolution

The growing standardization of pupillometric methods [13] [40], combined with established links to neuromodulatory systems [5] [65], positions pupillometry as an essential tool for advancing memory research beyond the limitations of behavioral measures alone.

Comparative Analysis of Pupillometry with Other Physiological and Neuroimaging Techniques

Within memory research, accurately measuring attentional allocation—a prerequisite for successful encoding and retrieval—remains a significant challenge. Pupillometry has emerged as a powerful, non-invasive tool for tracking cognitive processes in real-time through measurements of pupil diameter. These changes are linked to activity in brainstem arousal systems, most notably the Locus Coeruleus (LC), and the release of norepinephrine (NE), which regulates attention and cognitive control [5] [38]. This application note provides a comparative analysis of pupillometry against other established physiological and neuroimaging techniques. It details specific experimental protocols and reagents, offering a practical toolkit for researchers and drug development professionals to integrate these methods into studies of attention and memory.

Comparative Technique Analysis

The table below summarizes the key characteristics of pupillometry and other prominent techniques used in cognitive research.

Table 1: Comparison of Techniques for Assessing Attention and Memory Processes

Technique	Key Measured Parameters	Spatial Resolution	Temporal Resolution	Primary Applications in Memory/Attention Research	Key Advantages	Key Limitations
Pupillometry	Pupil diameter (tonic/phasic), constriction/ dilation velocity [5] [66].	Low	Very High (500+ Hz) [31]	Cognitive load, arousal, attentional engagement, "readiness to remember" [5] [67].	Non-invasive, cost-effective, high temporal resolution, excellent for real-time assessment [67].	Indirect neural measure; sensitive to luminance, eye movements, and autonomic state [31] [68].
fMRI	Blood-Oxygen-Level-Dependent (BOLD) signal.	High (mm-level)	Low (seconds)	Localizing brain activity to specific networks (e.g., FPN, Salience Network) during tasks [38].	Excellent spatial resolution for deep brain structures; whole-brain coverage.	Expensive, low temporal resolution, noisy environment restricts natural behavior.
Electroencephalography (EEG)	Event-Related Potentials (ERPs; e.g., P300, N100), oscillatory power [69].	Low	Very High (milliseconds)	Timing of cognitive processes; sensory and attentional processing stages [69].	Direct neural measure with millisecond temporal resolution; relatively low cost.	Poor spatial resolution; signal sensitive to muscle artifacts.
Eye-Tracking (Saccades/Fixations)	Gaze position, fixation duration, saccadic paths, time to first fixation [69].	Medium (1° visual angle)	High (250+ Hz)	Visual attention, information acquisition, attentional bias [69].	Direct measure of overt visual attention; easily integrated with pupillometry.	Cannot measure covert attention; data quality depends on stable head position.

Integrated Experimental Protocols

Protocol: Predicting Long-Term Memory Formation with Pre-Stimulus Pupillometry

This protocol is designed to investigate how pre-stimulus arousal states, indexed by baseline pupil size, influence subsequent memory performance, supporting the "Readiness to Remember" (R2R) framework [5].

• Objective: To determine if pupil diameter before and during stimulus encoding predicts successful long-term memory formation.
• Hypothesis: Successfully recalled or recognized stimuli will be associated with larger baseline and encoding-phase pupil diameters compared to forgotten stimuli.
• Materials & Setup:
  • Apparatus: An eye-tracker with a high-speed camera (≥ 500 Hz recommended) and infrared illumination to control for ambient light effects (e.g., EyeLink 1000 Plus) [31].
  • Stimulus Presentation Software: Software such as PsychoPy [31] or MATLAB for precise control of timing and stimulus display.
  • Stimuli: A set of 30 or more words (or images), matched for frequency, length, and semantic meaning [5].
  • Environment: A quiet, dimly lit room to minimize luminance fluctuations. Luminance of all visual stimuli must be carefully matched [31].
• Procedure:
  • Calibration: Perform a standard 9-point calibration and validation procedure for the eye-tracker.
  • Study Phase:
    • Each trial begins with a fixation cross (e.g., 1000 ms).
    • A baseline period follows, where the screen is neutral (e.g., 500 ms). Pupil size during this period is the pre-stimulus baseline.
    • A word is presented for encoding (e.g., 3000 ms). Pupil size is recorded throughout.
    • An inter-trial interval (e.g., 2000 ms) separates trials.
  • Distractor Task: A 5-minute arithmetic task to prevent rehearsal.
  • Test Phase: Administer either a free recall test (participants write down all remembered words) or a recognition test (participants identify "old" words among foils) [5].
• Data Analysis:
  • Pre-process pupil data: filter for blinks and artifacts.
  • Segment data into trials and extract mean pupil diameter during the baseline period and the stimulus encoding period.
  • For each participant, sort trials into "recalled" (Hit) and "forgotten" (Miss) based on test phase performance.
  • Use Bayesian or repeated-measures ANOVA to compare pupil size between recalled and forgotten trials during both baseline and encoding periods. A Bayesian Factor (BF10) > 10 is considered strong evidence for the alternative hypothesis [5].

Protocol: Simultaneous fMRI and Pupillometry to Dissociate Neural Networks

This protocol leverages the strengths of both fMRI and pupillometry to link cognitive pupillary responses to specific brain networks.

• Objective: To correlate pupil dynamics with BOLD activity in the Frontoparietal Network (FPN, related to cognitive load) and the Salience Network (related to stimulus relevance and autonomic arousal) during a working memory task [38].
• Hypothesis: Mean pupil size will correlate with FPN activity, while within-block changes in pupil size (pupil derivative) will correlate with Salience Network activity.
• Materials & Setup:
  • Apparatus: MRI scanner with a compatible, MR-safe infrared eye-tracking system (e.g., EyeLink MRI models).
  • Software: fMRI presentation software (e.g., Presentation, PsychToolbox) synchronized with the scanner.
  • Task: An N-back working memory task (e.g., 0-back, 1-back, 2-back) in a block design [38].
• Procedure:
  • Setup: Secure the pupil camera and ensure clear pupil capture. Perform calibration within the scanner bore using a calibration grid projected onto the screen.
  • Task Execution: Participants perform the N-back task while simultaneous fMRI and pupillometry data are collected.
  • Pupil Metrics:
    • Pupil Size: Mean pupil diameter during each task block.
    • Pupil Change: The first-order derivative of the pupil size signal, calculated within each block to capture rapid fluctuations [38].
• fMRI Analysis:
  • Preprocess fMRI data (realignment, normalization, smoothing).
  • Model two separate regressors in a general linear model (GLM):
    • Task blocks parametrically modulated by mean pupil size.
    • Task blocks parametrically modulated by the pupil change signal.
  • The first regressor will identify brain regions whose activity scales with overall cognitive load (FPN), while the second will identify regions involved in dynamic, stimulus-driven arousal (Salience Network) [38].

Signaling Pathways and Experimental Workflows

Neural Pathway of Cognitive Pupillary Response

The following diagram illustrates the central nervous system pathway that links cognitive effort to changes in pupil size.

Diagram Title: Neural Pathway of Cognitive Pupil Response

Integrated Pupillometry & EEG Experimental Workflow

This workflow charts the process of a simultaneous pupillometry and EEG study, as used in protocols investigating attentional bias [69].

Diagram Title: Pupillometry-EEG Integrated Study Workflow

The Scientist's Toolkit: Research Reagent Solutions

Table 2: Essential Materials and Tools for Integrated Pupillometry Research

Item Category	Specific Examples	Function & Application Notes
Pupillometry Hardware	Video-based eye trackers (EyeLink 1000 Plus), Tabletop/Handheld Pupillometers (NeurOptics NPi-200) [31] [66].	Measures pupil diameter and PLR with high precision. Tabletop models offer high accuracy for lab settings, while handheld pupillometers are suited for clinical/field use [70].
Stimulus Presentation Software	PsychoPy [31], MATLAB with PsychToolbox, E-Prime.	Prescribes experimental paradigms with millisecond precision, crucial for time-locking pupil and neural responses to stimuli.
Calibration Tools	9-point calibration grid, chin rest.	Ensures spatial accuracy of gaze and pupil data. A chin rest minimizes head movements and improves data quality [31].
Integrated Data Acquisition	EEG systems with integrated eye-tracking (e.g., from Brain Vision, EGI), fMRI-compatible eye trackers.	Enables synchronous recording of pupillary and neural (EEG/fMRI) data, allowing direct correlation of measures like P300 amplitude and pupil dilation [69].
Analysis Software & Algorithms	RIPA2 Algorithm [67], EyeTrack analysis software (e.g., SR Research Data Viewer), fMRI analysis packages (SPM, FSL).	Processes raw data. RIPA2 provides a real-time index of mental effort from pupil signals. Specialized software extracts fixation maps and BOLD correlates [38] [67].
Standardized Stimulus Sets	IAPS (for emotional stimuli), word lists (e.g., from MRC Psycholinguistic Database), novel/familiar food image sets [69].	Provides well-characterized, validated stimuli for memory and attention experiments, ensuring reproducibility and allowing cross-study comparisons.

Pupillometry, the measurement of pupil size and reactivity, has emerged as a critical tool in both clinical neurology and cognitive neuroscience research. The pupillary light reflex (PLR) represents a crucial biological reflex involving the retina, optic nerves, midbrain, brainstem, and both sympathetic and parasympathetic nervous systems [71]. Beyond its clinical utility in assessing neurological function, pupil size has been established as a reliable, non-invasive marker of cognitive processes such as attention, mental effort, and arousal through autonomic nervous system activity [72] [13]. Recent technological advancements have enabled the development of smartphone-based pupillometry platforms that leverage artificial intelligence to provide objective, quantitative measurements of pupil dynamics, making this technology more accessible for research settings including studies of attention and memory [71] [73].

Quantitative Validation of AI-Powered Pupillometry

Performance Metrics of Smartphone-Based Pupillometry

Table 1: Validation Metrics of SmartPLR Application Against Gold-Standard NPi-300

Parameter	SmartPLR Performance	Comparison Metric	Gold Standard Correlation
Segmentation Model	Mask R-CNN with ConvNeXt V2 backbone	—	—
Segmentation Accuracy	Mean Intersection over Union: 0.9177	—	—
Model Performance	Segmentation mAP: 0.8670; Bounding Box mAP: 0.8663	—	—
Pupil Size Difference	—	Pearson Correlation	0.77 [71]
Constriction Velocity (CV)	—	Pearson Correlation	0.77 [71]
Percentage Change in Pupil Size (CP)	—	Pearson Correlation	0.74 [71]
Pupil Reactivity Classification	SmartPLR Score Formula: (10^4 × pred_diff × pred_CV × (pred_CP)^2)	—	—

Performance in Cognitive Research Context

The reliability of smartphone-based pupillometry is particularly relevant for memory research, where pupil metrics serve as indicators of cognitive processing. Research has demonstrated that pupil size before and during stimulus presentation can predict long-term memory performance, with larger pupil diameter associated with successfully recalled words—a phenomenon supporting the "readiness to remember" framework [72]. This relationship between pupil-linked arousal and memory formation underscores the importance of precise measurement tools for advancing cognitive theory [72] [11].

Experimental Protocols

Protocol 1: Technical Validation of Smartphone Pupillometry

Objective: To validate the accuracy of smartphone-based pupillometry against the commercial infrared pupillometer (NPi-300) [71].

Materials: Samsung S20 Fe smartphone, NPi-300 Pupillometer (NeurOptics Inc.), dark room (<5 lx ambient light)

Procedure:

• Participant Preparation: Dark adaptation for 5 minutes before testing. Participants seated and instructed to look straight ahead. One eye tested at a time with the other occluded.
• Smartphone Data Acquisition:
  • Application identifies eye using MediaPipe acquisition algorithm.
  • Eye positioned in "trigger area" at center of screen (iris fills 0.325-0.5 times trigger area).
  • Automatic digital zoom (3x) activated when proper positioning achieved.
  • Recording: 1s pre-flash, 1.5s during flash, 0.5s post-flash at 30 Hz (1920×1080 resolution).
• Gold Standard Measurement:
  • After 5 minutes of dark adaptation, same eye measured with NPi-300.
  • Device captures 90 images within 3s with infrared camera after LED stimulus.
• Image Processing and Analysis:
  • Eye region cropped to 480×420 pixels.
  • CLAHE method applied to emphasize pupil edges.
  • Pupil and iris annotated using LabelMe tool for ground truth.
  • Deep learning models applied for segmentation.
  • Pupil-to-iris ratio analyzed rather than absolute measurements.

Protocol 2: Cognitive Pupillometry for Memory Research

Objective: To investigate the relationship between pupil metrics and memory encoding using smartphone pupillometry [72] [13].

Experimental Design Considerations:

• Stimulus Presentation: Ensure constant brightness between conditions as light is the main determinant of pupil size. Control for low-level differences beyond brightness that may affect pupil size [13].
• Trial Structure: Implement a trial-based design with baseline period (≥500ms) before stimulus onset to account for tonic pupil size [13].
• Data Collection Parameters: Sample pupil size at sufficient frequency (≥60Hz recommended) to capture task-evoked pupillary responses [13].

Procedure:

• Baseline Measurement: Record pupil size for 500-1000ms before stimulus presentation to establish baseline.
• Encoding Phase: Present memory stimuli while recording pupil dynamics.
• Retention Interval: Continue recording during retention period if investigating working memory.
• Recall/Recognition Test: Administer memory test after appropriate delay.
• Data Preprocessing:
  • Blink detection and interpolation.
  • Smoothing with band-pass filter (0.01-6Hz).
  • Baseline correction using pre-stimulus period.
  • Averaging across trials within conditions.

The Scientist's Toolkit: Essential Research Reagents and Materials

Table 2: Key Research Reagents and Materials for Smartphone Pupillometry Research

Item	Function/Application	Example/Specification
Smartphone Platform	Primary data acquisition device	Samsung S20 Fe (12MP ultrawide camera, f/2.2 aperture) [71]
Commercial Pupillometer	Gold standard validation	NPi-300 Pupillometer (NeurOptics Inc.) - captures 90 images in 3s with infrared [71]
AI Segmentation Models	Pupil detection and measurement	Mask R-CNN, UNet, UNet++, DeepLabV3, DeepLabV3+ [71]
Model Backbones	Feature extraction for segmentation	ResNet50, Swin Transformer, ConvNeXt V2 [71]
Image Annotation Tool	Ground truth generation	LabelMe tool for polygon annotation [71]
Computer Vision Library	Facial landmark detection	MediaPipe for eye identification and tracking [71]
Image Enhancement	Pupil edge emphasis	CLAHE method [71]
Specialized Software	Lighting-invariant pupil analysis	PuRe Pupillometer (Solvemed.ai) [73]

Workflow and System Architecture Visualization

Smartphone Pupillometry Data Acquisition Workflow

Cognitive Pupillometry Experimental Design

AI-Powered Pupil Segmentation Architecture

Implementation Considerations for Memory Research

When implementing smartphone pupillometry in memory research, several methodological factors require careful attention. The robustness of AI-powered segmentation for darker irises, historically challenging for visible-light systems, must be validated for specific research populations [71]. Environmental controls are essential, particularly for studies investigating the "readiness to remember" framework where baseline pupil size predicts memory performance [72]. Standardized protocols for baseline recording, stimulus presentation, and data preprocessing should be established across experiments to ensure comparability of findings [13].

The integration of these validated smartphone platforms enables new research paradigms in cognitive neuroscience, particularly for investigating how attentional states, as indexed by pupil metrics, influence memory encoding and retrieval. The accessibility of smartphone-based systems supports larger sample sizes and more diverse participant populations, potentially accelerating discovery in the cognitive neuroscience of memory.

Pupillometry, the precise measurement of pupil size and reactivity, has evolved beyond its traditional role in assessing basic arousal and physiological states. Within cognitive neuroscience, task-evoked pupil responses are recognized as a sensitive, non-invasive proxy for locus coeruleus-norepinephrine (LC-NE) system activity, which modulates attention, cognitive effort, and memory processes [13]. This physiological link provides a compelling rationale for its application in clinical trial contexts, particularly for evaluating therapeutics targeting cognitive domains. The pupil's response can serve as an objective quantitative biomarker for assessing drug effects on central nervous system function, especially where subjective cognitive assessments may be prone to bias or insensitivity. This document outlines specific application protocols and analytical frameworks for deploying pupillometry in pharmaceutical development, with a specific focus on its utility for measuring attention and memory-related cognitive load.

Application Notes: Key Use Cases in Drug Development

The integration of pupillometry into clinical trials offers distinct advantages for specific therapeutic areas and research questions. The following applications are supported by a growing body of methodological research.

Quantifying Cognitive Load and Listening Effort

Pupillometry is a well-validated objective measure of listening effort, particularly in studies involving hearing-impaired populations or pharmacological interventions aiming to reduce cognitive strain during auditory processing. Critical Insight: Pupil dilation can reveal subtle changes in cognitive demands even in the absence of performance differences on behavioral tasks, making it a sensitive marker for drug efficacy [74]. This is crucial for demonstrating a drug's effect on perceived effort, which may not be fully captured by accuracy or reaction time measures alone.

Assessing Novel Compounds Modulating the LC-NE System

Given the established relationship between pupil dynamics and norepinephrine release from the locus coeruleus, pupillometry is an ideal tool for evaluating compounds designed to modulate this pathway [13]. Drugs targeting conditions like attention-deficit/hyperactivity disorder (ADHD), Alzheimer's disease, or major depressive disorder can be assessed for their direct impact on this core neuromodulatory system. The task-evoked pupil response provides a translational biomarker that can be measured similarly in preclinical models and human trials.

Endpoints in Ophthalmologic Rare Disease Trials

Regulatory bodies are actively exploring novel efficacy endpoints for severe vision loss. The U.S. Food and Drug Administration, in a forthcoming 2025 workshop, is specifically evaluating the use of full-field stimulus threshold (FST) testing and other pupillometric measures to support regulatory decision-making for drugs and biologics [75]. This highlights the growing acceptance of pupillometry as a primary outcome in ophthalmologic drug development, particularly where standard visual acuity measures are insufficient.

Experimental Protocols

Standardized protocols are essential for generating reliable, reproducible data suitable for regulatory submissions. The following protocols are adapted from recent peer-reviewed methodologies.

Protocol 1: Rapid Clinical Pupillometry for Physiological Baseline

This protocol, adapted from Park et al. (2025), is designed for clinical settings to establish a baseline of autonomic and retinal function, which is critical for interpreting cognitive pupillometry results [76].

• Objective: To rapidly measure the pupillary light reflex (PLR) and establish baseline pupil parameters in a clinical trial population.
• Equipment: Portable, handheld pupillometer with capacity for full-field, chromatic light flashes.
• Stimuli:
  • Chromatic flashes (470 nm blue and 621 nm red).
  • Duration: 500 ms.
  • Retinal illuminance: 12,000 Td (troland).
• Dark Adaptation: 3 minutes prior to testing, as this duration was found to maximize PLR responses without impractically long delays [76].
• Procedure:
  • The participant is seated in a darkened room.
  • Following a 3-minute dark adaptation period, the pupillometer is positioned.
  • A series of light flashes (e.g., 4-6 trials per wavelength) are delivered to the eye.
  • A minimum 30-second interval is enforced between flashes to allow pupil recovery.
• Key Outcome Measures:
  • Baseline Pupil Size (BL): Mean size 1 second before flash onset.
  • Maximum Pupil Constriction (MPC): Peak constriction following the flash, often expressed as a percentage change from baseline.
  • Post-Illumination Pupillary Response (PIPR): Median pupil size 6-8 seconds after flash offset, a measure linked to melanopsin function.
• Data Quality Notes: Test-retest repeatability for this protocol is approximately 1 mm for BL and 10% for MPC and PIPR [76]. PIPR has been shown to be independent of age, making it a robust cross-sectional measure.

Protocol 2: Cognitive Load Assessment During a Memory Task

This protocol outlines a trial-based experiment to measure cognitive effort during a memory encoding or retrieval task.

• Objective: To quantify task-evoked pupil dilation as an index of cognitive strain or attentional allocation during a memory task.
• Equipment: Desktop or head-mounted eye-tracker with a high sampling rate (≥ 120 Hz).
• Stimuli & Task:
  • Participants perform a computerized memory task (e.g., word-list learning, n-back, or recognition memory test).
  • Stimuli are presented visually on a uniform, neutral-gray background.
  • Critical Design Control: The visual properties (luminance, contrast, spatial frequency) of all stimuli must be identical across experimental conditions to prevent light-driven pupil responses from confounding the cognitive measures [13].
• Procedure:
  • Each trial begins with a fixation cross (1000-2000 ms) to establish a pre-trial baseline.
  • A memory item (e.g., a word or image) is presented for encoding (e.g., 2000 ms).
  • This is followed by a retention interval and a retrieval probe.
  • The participant makes a behavioral response (e.g., "old/new" judgment).
  • The trial structure is repeated for multiple items (e.g., 100+ trials) to obtain a reliable signal.
• Key Outcome Measure: The task-evoked pupil response, calculated as the change in pupil diameter from the pre-stimulus baseline period to during and after stimulus presentation. The analysis focuses on the mean pupil diameter in a time window typically 500-1500 ms post-stimulus onset.

Analytical Framework and Baseline Correction

The choice of analysis method is paramount. A comparative study highlights that the start-time of the baseline period relative to stimulus onset can significantly influence conclusions, potentially capturing anticipatory cognitive processes like attention mobilization [74].

• Recommended Baseline Correction: The subtractive method is favored over divisive (percent change) methods, as the latter is sensitive to artifacts and very low baseline values [74].
  • Calculation: Baseline_Corrected_Pupil_Size = Raw_Pupil_Size - Mean_Baseline_Pupil_Size
• Normalization: To account for inter-individual differences in the dynamic range of pupil size, z-score transformation is recommended. This technique homogenizes variability across participants and conditions without altering the qualitative results [74].
  • Calculation: Z-Scored_Pupil_Size = (Raw_Pupil_Size - Mean_All_Trials) / Standard_Deviation_All_Trials

Data Presentation

Table 1: Comparison of Pupillometry Analysis Methods and Their Properties

Method	Calculation	Pros	Cons	Best Use Case
Subtractive Baseline Correction [74]	Raw - Mean_Baseline	Simple, less sensitive to low baseline values	Does not account for inter-individual dynamic range	Most cognitive pupillometry experiments; direct group comparisons
Divisive Baseline Correction [74]	(Raw - Baseline) / Baseline	Expresses change as a proportion	Highly sensitive to baseline artifacts and low values	Not generally recommended
Z-score Normalization [74]	(Raw - μ) / σ	Reduces inter-subject variability; allows direct comparison	Removes absolute size information	Individual differences research; combining data across subjects
Index of Cognitive Activity (ICA) [77]	Proprietary (wavelet-based)	Aims to isolate cognitive from light-related changes	Proprietary; not all eye trackers supported; mixed validation	Exploratory analysis in varying lighting
Index of Pupillary Activity (IPA) [77]	Fully transparent calculation	Open methodology; lifts software/hardware restrictions	Not consistently sensitive to cognitive strain	Alternative to ICA requiring transparent methods

Measure	Definition	Physiological Correlate	Test-Retest Repeatability	Age Correlation
Baseline Pupil Size (BL)	Mean diameter 1s before light flash	General arousal, autonomic tone	~1 mm	Trend to decrease with age (non-significant)
Maximum Pupil Constriction (MPC)	Peak constriction % after flash	Overall integrity of the PLR pathway	~10%	Trend to decrease with age (non-significant)
Post-Illumination Pupillary Response (PIPR)	Median size 6-8s after flash offset	Intrinsically photosensitive retinal ganglion cell (melanopsin) function	~10%	Independent of age

Signaling Pathways and Experimental Workflows

Cognitive Pupillometry Experimental Workflow

Neuromodulatory Pathway Linking Cognition to Pupil Dilation

The Scientist's Toolkit: Research Reagent Solutions

Table 3: Essential Materials and Analytical Tools for Cognitive Pupillometry

Item	Function / Rationale	Example / Note
High-Speed Eye Tracker	Records pupil size with sufficient temporal resolution to capture rapid dilations.	Sampling rate ≥ 120 Hz; ensure SDK provides raw pupil data.
Stimulus Presentation Software	Precisely controls timing and properties of visual/auditory stimuli.	Python (PsychoPy), MATLAB (Psychtoolbox), E-Prime.
Chromatic Light Source	For isolating specific retinal photoreceptor contributions in clinical PLR.	LEDs at 470 nm (melanopsin) and 621 nm (cone-weighted) [76].
Data Preprocessing Pipeline	Cleans raw data of blinks, artifacts, and smooths noise.	Use published code packages (e.g., in Python [13]).
Baseline Period Definition	Critical for calculating task-evoked change; captures pre-stimulus state.	Typically 200-1000 ms pre-stimulus; timing choice significantly impacts results [74].
Subtractive Baseline Correction	The recommended method for isolating cognitive pupil responses from baseline.	Simple subtraction of mean pre-stimulus diameter [74].
Z-score Transformation	Normalization technique to reduce inter-individual variability.	Homogenizes data across subjects for group-level analysis [74].

Conclusion

Pupillometry has firmly established itself as a robust, non-invasive window into the cognitive processes of attention and memory. The synthesis of research confirms that pupil size, both at baseline and in response to tasks, is a reliable predictor of memory encoding success, grounded in the activity of the LC-NE system. Adherence to rigorous methodological standards is paramount for data integrity, particularly in baseline selection and artifact management. The validation of pupillary measures against behavioral outcomes and their sensitivity to cognitive effort, even in the absence of performance differences, underscores their unique value. Future directions should focus on standardizing protocols across labs, further exploring the diagnostic and prognostic utility of pupillometry in neurodegenerative diseases, and leveraging emerging AI-driven and mobile technologies to transform cognitive assessment in both clinical and pharmaceutical development settings.

References

• 1. norepinephrine system to the coupling between pupil-linked ... [https://pubmed.ncbi.nlm.nih.gov/40463220/]
• 2. norepinephrine system to the coupling between pupil ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12132287/]
• 3. Rational regulation of learning dynamics by pupil–linked ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC3386464/]
• 4. Pupil-Linked Arousal Modulates Precision of Stimulus ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11484544/]
• 5. Predicting long-term memory via pupillometry [https://www.nature.com/articles/s41598-025-09703-7]
• 6. Pupil-linked arousal signals track the temporal ... - Nature [https://www.nature.com/articles/s41467-020-17851-9]
• 7. Pupil-Linked Arousal Determines Variability in Perceptual ... [https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1003854]
• 8. Pupil size reflects successful encoding and recall of ... [https://www.nature.com/articles/s41598-018-23197-6]
• 9. Dissociating Spatial Attention and Working Memory ... [https://pubmed.ncbi.nlm.nih.gov/41066699/]
• 10. Predicting long-term memory via pupillometry - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC12217946/]
• 11. Investigating a mental effort explanation of the generation ... [https://link.springer.com/article/10.3758/s13421-025-01791-0]
• 12. Quantitative pupillometry as a sensitive biomarker for ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12488747/]
• 13. Methods in cognitive pupillometry: Design, preprocessing, and ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC10556184/]
• 14. Quantitative determination of pupil by dynamic pupillometry ... [https://www.sciencedirect.com/science/article/pii/S221339842030018X]
• 15. Pupillometric Complexity and Symmetricity Follow Inverted- ... [https://www.frontiersin.org/journals/physiology/articles/10.3389/fphys.2021.614479/full]
• 16. Pupillometry as a Window into Young Children's Sustained ... [https://www.mdpi.com/2079-3200/10/4/107]
• 17. Pupillometry signatures of sustained attention and working ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC10158816/]
• 18. Methods in cognitive pupillometry: Design, preprocessing ... [https://link.springer.com/article/10.3758/s13428-022-01957-7]
• 19. Eyes on Memory: Pupillometry in Encoding and Retrieval [https://www.mdpi.com/2411-5150/8/2/37]
• 20. The “inverted-U” of cognitive effort across speech rates ... [https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2025.1685938/full]
• 21. The “inverted-U” of cognitive effort across speech rates holds ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12614458/]
• 22. Focused attention predicts visual working memory ... [https://www.sciencedirect.com/science/article/pii/S1878929318300409]
• 23. Pupillometry as a biomarker of postural control [https://pmc.ncbi.nlm.nih.gov/articles/PMC11579225/]
• 24. Eyes on Memory: Pupillometry in Encoding and Retrieval - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC11209248/]
• 25. The pupillometric production effect: Evidence for enhanced ... [https://www.sciencedirect.com/science/article/pii/S0010027725002677]
• 26. Establishing a normative database for quantitative ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC7098071/]
• 27. Colorful glares: Effects of colors on brightness illusions ... [https://www.sciencedirect.com/science/article/pii/S0001691819300083]
• 28. Racially fair pupillometry measurements for RGB ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC10449795/]
• 29. Pupillometry: Psychology, Physiology, and Function - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC6634360/]
• 30. Is Pupil Activity Associated With the Strength of Memory ... [https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2021.686183/full]
• 31. Luminance Matching in Cognitive Pupillometry Is Not Enough [https://www.eneuro.org/content/12/11/ENEURO.0238-25.2025]
• 32. Table 3 Quantitative pupillometry data reported by previous ... [https://bmcophthalmol.biomedcentral.com/articles/10.1186/s12886-020-01389-x/tables/3]
• 33. SPEED: A Graphical User Interface Software for Processing ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12015838/]
• 34. Attachment is in the eye of the beholder: a pupillometry ... [https://www.nature.com/articles/s41598-025-92347-4]
• 35. Analyzing the Time Course of Pupillometric Data - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC6535748/]
• 36. Keeping an Eye on Effort: A Pupillometric Investigation of ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC7809337/]
• 37. Evaluating the learning of stimulus-control associations ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC8758512/]
• 38. Pupillometry tracks cognitive load and salience network ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC8720183/]
• 39. Best Practices and Advice for Using Pupillometry to Measure ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC6166306/]
• 40. PupilMetrics: a support system for preprocessing of ... [https://www.nature.com/articles/s41598-024-79920-z]
• 41. Establishing a normative database for quantitative ... [https://bmcophthalmol.biomedcentral.com/articles/10.1186/s12886-020-01389-x]
• 42. eyeris: A flexible, extensible, and reproducible pupillometry ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12157406/]
• 43. Preprocessing pupil size data: Guidelines and code [https://link.springer.com/article/10.3758/s13428-018-1075-y]
• 44. A Preprocessing Pipeline for Pupillometry Signal from ... [https://www.mdpi.com/1424-8220/25/15/4737]
• 45. Dimensionality reduction techniques in pupillometry research [https://link.springer.com/article/10.3758/s13428-025-02786-0]
• 46. 2.4. Pupillometry analysis [https://bio-protocol.org/exchange/minidetail?id=17742601&type=30]
• 47. A rapid pupillometry protocol for clinical use: effect of age ... [https://link.springer.com/article/10.1007/s10633-025-10068-5]
• 48. A Preprocessing Pipeline for Pupillometry Signal from ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12349379/]
• 49. Attention Cueing in Rivalry: Insights from Pupillometry [https://www.eneuro.org/content/9/3/ENEURO.0497-21.2022]
• 50. Systematic Comparison of Trial Exclusion Criteria for ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC8150669/]
• 51. Preprocess • pupillometry [https://dr-jt.github.io/pupillometry/articles/preprocess_pupil.html]
• 52. Safe and sensible preprocessing and baseline correction ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC5809553/]
• 53. PupilMetrics: a support system for preprocessing of ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11579351/]
• 54. Prestimulus subsequent memory effects for auditory and visual events - PMC [https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3881065/]
• 55. Pupillometry as tool to assess cognitive and affective ... [https://www.sciencedirect.com/science/article/pii/S2666459324000143]
• 56. Neurophysiological signature of effective anticipatory task-set control: a task-switching investigation - PubMed [https://pubmed.ncbi.nlm.nih.gov/18717737/]
• 57. Chromatic Pupillometry Methods for Assessing Photoreceptor ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC6379484/]
• 58. Handheld chromatic pupillometry can accurately and rapidly ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC10176376/]
• 59. Give me a break! Unavoidable fatigue effects in cognitive ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11161670/]
• 60. Pupillometry as a window to detect cognitive aging in the ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC10770000/]
• 61. Pupillometry in Critical Care | Measure Pupil Size [https://neuroptics.com/pupillometry-clinical-overview/]
• 62. First steps into the pupillometry multiverse of developmental ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11133157/]
• 63. Uncovering Distinct Drivers of Covert Attention in Complex ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11920946/]
• 64. Phonological discrimination and contrast detection in ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC10646334/]
• 65. evidence from pupillometry, fMRI, and RNN Modelling [https://elifesciences.org/reviewed-preprints/93191v1]
• 66. Objective Pupillometry as an Adjunct to Prediction and Assessment for Oculomotor Nerve Injury and Recovery: Potential for Practical Applications - ScienceDirect [https://www.sciencedirect.com/science/article/abs/pii/S1878875018321867]
• 67. Measuring Mental Effort in Real Time Using Pupillometry [https://www.mdpi.com/1995-8692/18/6/70]
• 68. Too Bright to Focus? Influence of Brightness Illusions and ... [https://www.mdpi.com/2411-5150/9/4/81]
• 69. Exploring the Link Between Visual Attention to Familiar or ... [https://www.researchprotocols.org/2025/1/e69541]
• 70. Pupillometer Market Analysis, Size, and Forecast 2025-2029 [https://www.technavio.com/report/pupillometer-market-industry-analysis]
• 71. a digital solution for AI-powered smartphone pupillometry [https://bmcophthalmol.biomedcentral.com/articles/10.1186/s12886-025-04462-5]
• 72. Predicting long-term memory via pupillometry [https://pubmed.ncbi.nlm.nih.gov/40595394/]
• 73. Solvemed.ai ― Objective pupil reactivity testing. Now in your ... [https://www.solvemed.ai/]
• 74. Comparing methods of analysis in pupillometry [https://pmc.ncbi.nlm.nih.gov/articles/PMC9207619/]
• 75. Efficacy Endpoints in Ophthalmologic Rare Disease Drug ... [https://www.fda.gov/drugs/news-events-human-drugs/assessing-novel-efficacy-endpoints-ophthalmologic-rare-disease-drug-and-biologics-development]
• 76. A rapid pupillometry protocol for clinical use: effect of age ... [https://pubmed.ncbi.nlm.nih.gov/41240211/]
• 77. Comparison of two pupillometric measures as indicators ... [https://www.sciencedirect.com/science/article/pii/S000368702400019X]