Ami Eidels
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Ami Eidels has a total of 118 co-authors. Here are the top co-authors by number of shared publications:
- Daniel R. Little
Melbourne School of Psychological Sciences, The University of Melbourne, Melbourne, Australia
12 shared publications - Scott D. Brown
School of Psychological Science, College of Engineering, Science and Environment, University of Newcastle, Callaghan 2308, Australia
12 shared publications - Zachary L. Howard
A Population Health, Hunter New England Local Health District, Newcastle, NSW, Australia
12 shared publications - James T. Townsend
Indiana University
11 shared publications - Joseph Woodworth Houpt
Department of Psychology, Wright State University, Dayton, OH, USA
10 shared publications - Andrew Heathcote
Department of Psychology, University of Amsterdam, Amsterdam, The Netherlands
9 shared publications
Latest Publications
- Open Access
- 27 October 2026
- Published by Center for Open Science in PsyArXiv
CogFlow: Open-Source Solution for the Construction of Modular Cognitive Psychology Experiments
Proliferating web-based experimentation platforms have substantially broadened access to cognitive psychology research infrastructure, yet no existing tool simultaneously supports continuous and trial-based paradigms, maintains institutional data-protection compliance, works across online and laboratory settings, and operates within the constraints of collaborative multi-site teams. We present CogFlow, a free, open-source, self-hostable web platform for constructing, deploying, and managing cognitive psychology experiments. CogFlow’s JSON-centric architecture uses a single declarative configuration file to specify both study design and runtime instructions for the Interpreter engine, supporting a broad task library including Random Dot Motion, Flanker, SART, Stroop and Emotional Stroop, Simon, Task Switching, PVT, MOT, N-back, Gabor patch, Continuous Image Presentation, and Security Operations Center (SOC)-style multitask environments, in both trial-based and continuous-presentation modes. A no-code graphical Builder lowers the barrier to experiment authoring while preserving full parameter-level configurability for researchers who require it; results are collected as encrypted payloads with multi-factor authentication gating and pseudonymized participant linkage, supporting compliance with GDPR, UK GDPR, and IRB data management requirements. The platform integrates natively with SONA Systems and Prolific Academic for participant recruitment, provides in-portal descriptive analysis and charting, and includes a CogFlow To Go packaging mode for offline laboratory deployment with Lab Streaming Layer (LSL) support for physiological co-registration. Software-layer runtime benchmarks confirm stable performance consistent with the broader literature on browser-based platforms. CogFlow is available at https://cogflow.app under the BSD 3-Clause license; contributor roles are tracked using the CRediT taxonomy with versioned archival via Zenodo.
- 27 October 2026
- Published by Center for Open Science in PsyArXiv
Open AccessCogFlow: Open-Source Solution for the Construction of Modular Cognitive Psychology Experiments
Proliferating web-based experimentation platforms have substantially broadened access to cognitive psychology research infrastructure, yet no existing tool simultaneously supports continuous and trial-based paradigms, maintains institutional data-protection compliance, works across online and laboratory settings, and operates within the constraints of collaborative multi-site teams. We present CogFlow, a free, open-source, self-hostable web platform for constructing, deploying, and managing cognitive psychology experiments. CogFlow’s JSON-centric architecture uses a single declarative configuration file to specify both study design and runtime instructions for the Interpreter engine, supporting a broad task library including Random Dot Motion, Flanker, SART, Stroop and Emotional Stroop, Simon, Task Switching, PVT, MOT, N-back, Gabor patch, Continuous Image Presentation, and Security Operations Center (SOC)-style multitask environments, in both trial-based and continuous-presentation modes. A no-code graphical Builder lowers the barrier to experiment authoring while preserving full parameter-level configurability for researchers who require it; results are collected as encrypted payloads with multi-factor authentication gating and pseudonymized participant linkage, supporting compliance with GDPR, UK GDPR, and IRB data management requirements. The platform integrates natively with SONA Systems and Prolific Academic for participant recruitment, provides in-portal descriptive analysis and charting, and includes a CogFlow To Go packaging mode for offline laboratory deployment with Lab Streaming Layer (LSL) support for physiological co-registration. Software-layer runtime benchmarks confirm stable performance consistent with the broader literature on browser-based platforms. CogFlow is available at https://cogflow.app under the BSD 3-Clause license; contributor roles are tracked using the CRediT taxonomy with versioned archival via Zenodo.
- Open Access
- 22 October 2026
- Published by Center for Open Science in PsyArXiv
Prioritization of Perceptual Decisions
Scheduling theory provides optimal policies for allocating limited resources to multiple tasks over time. However, little research has examined how cognitive mechanisms perform relative to these optimal policies. We introduce a novel experimental paradigm to study human scheduling decisions using a set of perceptual tasks (random dot kinematograms, RDKs) that vary in difficulty. A foundational question in this paradigm is how optimal prioritization depends on the representation of task factors. Participants selected and completed RDKs one at a time, with the goal of completing as many as possible before a deadline. In Experiment 1 (collected in 2020), RDK difficulty was explicitly labeled. Participants showed near-optimal scheduling, selecting easier tasks first, especially under time pressure. In Experiment 2 (collected in 2022), difficulty had to be inferred from dynamic RDK displays. This resulted in less optimal scheduling, closer to random selection. In Experiment 3 (collected in 2026), we removed observation time from the selection and completion time by limiting the time window in which participants could view the moving RDKs. Across increasing time windows from 200 to 1600 ms, we showed that the optimality of the resulting schedule grew with increasing observation time. We develop a cognitive model of scheduling based on the Plackett-Luce model of ranked preferences. The model captures key differences between experiments in terms of the strength parameters associated with each difficulty level. Our paradigm and modeling approach provide a foundation for studying human scheduling across a range of task types and goals.
- 22 October 2026
- Published by Center for Open Science in PsyArXiv
Open AccessPrioritization of Perceptual Decisions
Scheduling theory provides optimal policies for allocating limited resources to multiple tasks over time. However, little research has examined how cognitive mechanisms perform relative to these optimal policies. We introduce a novel experimental paradigm to study human scheduling decisions using a set of perceptual tasks (random dot kinematograms, RDKs) that vary in difficulty. A foundational question in this paradigm is how optimal prioritization depends on the representation of task factors. Participants selected and completed RDKs one at a time, with the goal of completing as many as possible before a deadline. In Experiment 1 (collected in 2020), RDK difficulty was explicitly labeled. Participants showed near-optimal scheduling, selecting easier tasks first, especially under time pressure. In Experiment 2 (collected in 2022), difficulty had to be inferred from dynamic RDK displays. This resulted in less optimal scheduling, closer to random selection. In Experiment 3 (collected in 2026), we removed observation time from the selection and completion time by limiting the time window in which participants could view the moving RDKs. Across increasing time windows from 200 to 1600 ms, we showed that the optimality of the resulting schedule grew with increasing observation time. We develop a cognitive model of scheduling based on the Plackett-Luce model of ranked preferences. The model captures key differences between experiments in terms of the strength parameters associated with each difficulty level. Our paradigm and modeling approach provide a foundation for studying human scheduling across a range of task types and goals.
- Open Access
- 12 August 2025
- Published by Center for Open Science in PsyArXiv
Explaining Multi-Scale Choice Dynamics
Sequences of choice response times exhibit ubiquitous and strong multi-scale dynamics (i.e., sequential dependencies across a broad range of temporal scales). Despite their pervasive nature, multi-scale dynamics are poorly understood. We show that dynamics in the seconds to minutes range can be explained by the superposition of several distinct learning and control mechanisms. Each mechanism learns a representation of the structure of the choice environment and/or an aspect of the decision maker’s ability. These representations are updated after each choice and used to control the next decision by modulating the parameters of an evidence-accumulation process with sub-second range dynamics that determine individual choices. We link these mechanisms to three major foci in the experimental study of sequential dependencies: stimulus-history, error-related, and hard-easy effects. This account provides a detailed explanation of both multi-scale dynamics of choice sequences, and the three effects, at the group and individual levels.
- 12 August 2025
- Published by Center for Open Science in PsyArXiv
Open AccessExplaining Multi-Scale Choice Dynamics
Sequences of choice response times exhibit ubiquitous and strong multi-scale dynamics (i.e., sequential dependencies across a broad range of temporal scales). Despite their pervasive nature, multi-scale dynamics are poorly understood. We show that dynamics in the seconds to minutes range can be explained by the superposition of several distinct learning and control mechanisms. Each mechanism learns a representation of the structure of the choice environment and/or an aspect of the decision maker’s ability. These representations are updated after each choice and used to control the next decision by modulating the parameters of an evidence-accumulation process with sub-second range dynamics that determine individual choices. We link these mechanisms to three major foci in the experimental study of sequential dependencies: stimulus-history, error-related, and hard-easy effects. This account provides a detailed explanation of both multi-scale dynamics of choice sequences, and the three effects, at the group and individual levels.
- Open Access
- 10 February 2025
- Published by Center for Open Science in PsyArXiv
Dividing up the pie: allocating attention across dual-tasks
Daily life, whether at home or at work, requires people to split their attention. These simultaneous demands can be exhausting and one strategy is to reallocate attention towards a pressing demand at the expense of others. Choosing how to allocate limited cognitive resources is the basis of many psychological theories but few have directly investigated the cognitive mechanism that drives the resource allocation process. In this paper, we present a novel joint-computational approach that highlights the central role of attentional resource allocation in dual-task environments. We present two paired experiments where participants simultaneously complete a Multiple-Object-Tracking task (MOT) alongside a Detection Response Task (DRT). In breaking with previous modeling investigations, our computational framework unites separate decision-making models under the control of a single allocation process that makes accurate predictions under a range of cognitive demands. Foremost, the current investigation provides a framework for quantifying the subjective and latent processes involved in divided attention. We discuss the theoretical implications of our approach and highlight the flexibility of decision-makers to adjust where resources are spent under dynamically shifting workloads.
- 10 February 2025
- Published by Center for Open Science in PsyArXiv
Open AccessDividing up the pie: allocating attention across dual-tasks
Daily life, whether at home or at work, requires people to split their attention. These simultaneous demands can be exhausting and one strategy is to reallocate attention towards a pressing demand at the expense of others. Choosing how to allocate limited cognitive resources is the basis of many psychological theories but few have directly investigated the cognitive mechanism that drives the resource allocation process. In this paper, we present a novel joint-computational approach that highlights the central role of attentional resource allocation in dual-task environments. We present two paired experiments where participants simultaneously complete a Multiple-Object-Tracking task (MOT) alongside a Detection Response Task (DRT). In breaking with previous modeling investigations, our computational framework unites separate decision-making models under the control of a single allocation process that makes accurate predictions under a range of cognitive demands. Foremost, the current investigation provides a framework for quantifying the subjective and latent processes involved in divided attention. We discuss the theoretical implications of our approach and highlight the flexibility of decision-makers to adjust where resources are spent under dynamically shifting workloads.
- Open Access
- 20 December 2023
- Published by The Royal Society in Royal Society Open Science
- Vol. 10 (12) , 231613
Prior beliefs and the interpretation of scientific results
How do prior beliefs affect the interpretation of scientific results? I discuss a hypothetical scenario where researchers publish results that could either support a theory they believe in, or refute that theory, and ask if the two instances carry the same weight. More colloquially, I ask if we should overweigh scientific results supporting a given theory and reported by a researcher, or a team, that initially did not support that theory. I illustrate the challenge using two examples from psychology: evidence accumulation models, and extra sensory perception.
- 20 December 2023
- Published by The Royal Society in Royal Society Open Science
- Vol. 10 (12) , 231613
Open AccessPrior beliefs and the interpretation of scientific results
How do prior beliefs affect the interpretation of scientific results? I discuss a hypothetical scenario where researchers publish results that could either support a theory they believe in, or refute that theory, and ask if the two instances carry the same weight. More colloquially, I ask if we should overweigh scientific results supporting a given theory and reported by a researcher, or a team, that initially did not support that theory. I illustrate the challenge using two examples from psychology: evidence accumulation models, and extra sensory perception.
- Open Access
- 30 August 2023
- Published by Elsevier in Journal of Choice Modelling
- Vol. 49
Do choice tasks and rating scales elicit the same judgments?
- 30 August 2023
- Published by Elsevier in Journal of Choice Modelling
- Vol. 49
Open AccessDo choice tasks and rating scales elicit the same judgments?
- Open Access
- 21 July 2023
- Published by Center for Open Science in PsyArXiv
Trust in Human-bot Teaming: Applications of the Judge Advisor System
Recent years have seen remarkable advances in the development and use of Artificial Intelligence (AI) in image classification, driving cars, and writing scientific articles. Although AI can outperform humans in many tasks, there remain domains where humans and AI working together can outperform either working alone. For humans and AI to work together effectively, the human must trust the AI bot to the right degree (calibrated). If the human does not trust the bot sufficiently, or conversely trusts the bot more than is warranted, the human-bot team will not perform as well as they could. We report three experiments examining trust in human-AI teaming. While existing studies typically collect binary responses (to trust, or not to trust), we present a novel paradigm that quantifies trust in a bot-recommendation in a continuous fashion. These data allow better precision, and in the future the development of more refined models of human-bot trust.
- 21 July 2023
- Published by Center for Open Science in PsyArXiv
Open AccessTrust in Human-bot Teaming: Applications of the Judge Advisor System
Recent years have seen remarkable advances in the development and use of Artificial Intelligence (AI) in image classification, driving cars, and writing scientific articles. Although AI can outperform humans in many tasks, there remain domains where humans and AI working together can outperform either working alone. For humans and AI to work together effectively, the human must trust the AI bot to the right degree (calibrated). If the human does not trust the bot sufficiently, or conversely trusts the bot more than is warranted, the human-bot team will not perform as well as they could. We report three experiments examining trust in human-AI teaming. While existing studies typically collect binary responses (to trust, or not to trust), we present a novel paradigm that quantifies trust in a bot-recommendation in a continuous fashion. These data allow better precision, and in the future the development of more refined models of human-bot trust.
- Open Access
- 1 September 2022
- Published by Center for Open Science in PsyArXiv
Human Performance in Competitive and Collaborative Human-Machine Teams
In the modern world, there are important tasks that have become too complex for a single unaided individual to manage. Some safety-critical tasks are conducted by teams to improve task performance and minimize risk of error. These teams have traditionally consisted of human operators, yet nowadays AI and machine systems are incorporated into team environments to improve performance and capacity. We used a computerized task, modeled after a classic arcade game, to investigate the performance of human-machine and human-human teams. We manipulated the group conditions between team members; sometimes they were incentivised to collaborate, sometimes compete, and sometimes to work separately. We evaluated players’ performance in the main task (game play) and also measured the cognitive workload they experienced. We compared workload and game performance between different team types (human-human vs. human-machine) and different group conditions (competitive, collaborate, independent). Adapting workload capacity analysis to human-machine teams, we found performance under both team types and all group conditions suffered a performance efficiency cost. However, we observed a reduced cost in collaborative over competitive teams within human-human pairings but this effect was diminished when playing with a machine partner. The implications of workload capacity analysis as a powerful tool for human-machine team performance measurement are discussed.
- 1 September 2022
- Published by Center for Open Science in PsyArXiv
Open AccessHuman Performance in Competitive and Collaborative Human-Machine Teams
In the modern world, there are important tasks that have become too complex for a single unaided individual to manage. Some safety-critical tasks are conducted by teams to improve task performance and minimize risk of error. These teams have traditionally consisted of human operators, yet nowadays AI and machine systems are incorporated into team environments to improve performance and capacity. We used a computerized task, modeled after a classic arcade game, to investigate the performance of human-machine and human-human teams. We manipulated the group conditions between team members; sometimes they were incentivised to collaborate, sometimes compete, and sometimes to work separately. We evaluated players’ performance in the main task (game play) and also measured the cognitive workload they experienced. We compared workload and game performance between different team types (human-human vs. human-machine) and different group conditions (competitive, collaborate, independent). Adapting workload capacity analysis to human-machine teams, we found performance under both team types and all group conditions suffered a performance efficiency cost. However, we observed a reduced cost in collaborative over competitive teams within human-human pairings but this effect was diminished when playing with a machine partner. The implications of workload capacity analysis as a powerful tool for human-machine team performance measurement are discussed.
- Open Access
- 30 August 2022
- Published by Center for Open Science in PsyArXiv
The Relationship Between Teaming Behaviours and Joint Capacity of Hybrid Human-Machine Teams
Collaboration in shared environments requires human agents to coordinate their behaviour according to the machines’ actions. In this study, we compared the performance and behaviour of Human-Machine (HM) and Human-Human (HH) teams. While HH teaming behaviour is sensitive to Collaborative contexts, little is known about HM teaming behaviour. Furthermore, teaming behaviour may impact the team’s Joint Capacity – the team’s ability to handle teamwork processes and task demands. To assess teaming behaviour at every moment of a trial we used three distinct spatiotemporal measures (Momentary Distance, Highly Correlated Segments, and Running Correlation). To assess the team’s joint performance, we adopted the Capacity Coefficient (Townsend & Nozawa,1995). For both HH and HM teams, behavioural measures predicted Joint Capacity. HH teams demonstrated greater performance and less synchronous behaviour than HM teams. The reduced synchrony of HH teams likely improved their performance as they could complement each other’s behaviour ratherthan duplicate inefficiencies
- 30 August 2022
- Published by Center for Open Science in PsyArXiv
Open AccessThe Relationship Between Teaming Behaviours and Joint Capacity of Hybrid Human-Machine Teams
Collaboration in shared environments requires human agents to coordinate their behaviour according to the machines’ actions. In this study, we compared the performance and behaviour of Human-Machine (HM) and Human-Human (HH) teams. While HH teaming behaviour is sensitive to Collaborative contexts, little is known about HM teaming behaviour. Furthermore, teaming behaviour may impact the team’s Joint Capacity – the team’s ability to handle teamwork processes and task demands. To assess teaming behaviour at every moment of a trial we used three distinct spatiotemporal measures (Momentary Distance, Highly Correlated Segments, and Running Correlation). To assess the team’s joint performance, we adopted the Capacity Coefficient (Townsend & Nozawa,1995). For both HH and HM teams, behavioural measures predicted Joint Capacity. HH teams demonstrated greater performance and less synchronous behaviour than HM teams. The reduced synchrony of HH teams likely improved their performance as they could complement each other’s behaviour ratherthan duplicate inefficiencies
- Open Access
- 9 August 2022
- Published by Center for Open Science in PsyArXiv
Do Choice Tasks and Rating Scales Elicit the Same Judgments?
Discrete choice (DCE) and rating scale experiments (RSE) are commonly applied procedures for eliciting preference judgments in a plethora of applied settings such as consumer choices, health care, and transport economics. An almost universal assumption is that actual "ground truth" preferences do not depend on which elicitation procedure is used. It is usually not possible to test this assumption, because typical studies feature response options for which there is no objectively correct response. To make progress on testing this assumption, we conducted a perceptual discrimination experiment where response options varied on a single attribute -- stimulus saturation level -- with a known objectively correct response. We had the same participants complete both a choice task (CT) and rating scale (RS) version of the experiment, allowing a direct examination of the assumption of a common representation. Our CT featured many characteristics that define a DCE, however, in order to have a known objectively correct response, it also differed in a few important ways. To test the assumption of a common representation, we developed a cognitive model with a response mechanism for both CT and RS. This enabled us to compare a model version that featured one shared latent stimulus representation across CT and RS versus a version which featured separate representations. Our results support the assumption that a single internal state supports both CT and RS responses, and also suggest that the CT method might provide more sensitive measurement of internal states than the RS method.
- 9 August 2022
- Published by Center for Open Science in PsyArXiv
Open AccessDo Choice Tasks and Rating Scales Elicit the Same Judgments?
Discrete choice (DCE) and rating scale experiments (RSE) are commonly applied procedures for eliciting preference judgments in a plethora of applied settings such as consumer choices, health care, and transport economics. An almost universal assumption is that actual "ground truth" preferences do not depend on which elicitation procedure is used. It is usually not possible to test this assumption, because typical studies feature response options for which there is no objectively correct response. To make progress on testing this assumption, we conducted a perceptual discrimination experiment where response options varied on a single attribute -- stimulus saturation level -- with a known objectively correct response. We had the same participants complete both a choice task (CT) and rating scale (RS) version of the experiment, allowing a direct examination of the assumption of a common representation. Our CT featured many characteristics that define a DCE, however, in order to have a known objectively correct response, it also differed in a few important ways. To test the assumption of a common representation, we developed a cognitive model with a response mechanism for both CT and RS. This enabled us to compare a model version that featured one shared latent stimulus representation across CT and RS versus a version which featured separate representations. Our results support the assumption that a single internal state supports both CT and RS responses, and also suggest that the CT method might provide more sensitive measurement of internal states than the RS method.