Alejandro Tlaie
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Alejandro Tlaie has a total of 59 co-authors. Here are the top co-authors by number of shared publications:
- Katharine Shapcott
Ernst Strüngmann Institute for Neuroscience in cooperation with the Max Planck Society, Frankfurt am Main, Germany
5 shared publications Berkutay Mert
Ernst Strüngmann Institute for Neuroscience in cooperation with the Max Planck Society, Frankfurt am Main, Germany
3 shared publications- Jonathan W. Pillow
Princeton Neuroscience Institute, Princeton University
3 shared publications - Martha Nari Havenith
Zero-Noise Lab, Ernst Strungmann Institute of the Max Planck Society, Frankfurt a.M., Germany
3 shared publications Mina Glukhova
Ernst Strüngmann Institute for Neuroscience in cooperation with the Max Planck Society, Frankfurt am Main, Germany
3 shared publications- Muad Y. Abd El Hay
Zero-Noise Lab, Ernst Strungmann Institute of the Max Planck Society, Frankfurt a.M., Germany
3 shared publications
Latest Publications
- Open Access
- 25 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 5095-5113
Expanding External Access to Frontier AI Models for Dangerous Capability Evaluations
- 25 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 5095-5113
Open AccessExpanding External Access to Frontier AI Models for Dangerous Capability Evaluations
- Open Access
- 4 June 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-16
Inferring internal states across mice and monkeys using facial features
Animal behaviour is shaped to a large degree by internal cognitive states, but it is unknown whether these states are similar across species. To address this question, here we develop a virtual reality setup in which male mice and macaques engage in the same naturalistic visual foraging task. We exploit the richness of a wide range of facial features extracted from video recordings during the task, to train a Markov-Switching Linear Regression (MSLR). By doing so, we identify, on a single-trial basis, a set of internal states that reliably predicts when the animals are going to react to the presented stimuli. Even though the model is trained purely on reaction times, it can also predict task outcome, supporting the behavioural relevance of the inferred states. The relationship of the identified states to task performance is comparable between mice and monkeys. Furthermore, each state corresponds to a characteristic pattern of facial features that partially overlaps between species, highlighting the importance of facial expressions as manifestations of internal cognitive states across species.
- 4 June 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-16
Open AccessInferring internal states across mice and monkeys using facial features
Animal behaviour is shaped to a large degree by internal cognitive states, but it is unknown whether these states are similar across species. To address this question, here we develop a virtual reality setup in which male mice and macaques engage in the same naturalistic visual foraging task. We exploit the richness of a wide range of facial features extracted from video recordings during the task, to train a Markov-Switching Linear Regression (MSLR). By doing so, we identify, on a single-trial basis, a set of internal states that reliably predicts when the animals are going to react to the presented stimuli. Even though the model is trained purely on reaction times, it can also predict task outcome, supporting the behavioural relevance of the inferred states. The relationship of the identified states to task performance is comparable between mice and monkeys. Furthermore, each state corresponds to a characteristic pattern of facial features that partially overlaps between species, highlighting the importance of facial expressions as manifestations of internal cognitive states across species.
- 25 April 2025
- Published by Springer Nature
- p. 420-442
Exploring and Steering the Moral Compass of Large Language Models
- 25 April 2025
- Published by Springer Nature
- p. 420-442
Exploring and Steering the Moral Compass of Large Language Models
- Open Access
Securing External Deeper-than-black-box GPAI Evaluations
This paper examines the critical challenges and potential solutions for conducting secure and effective external evaluations of general-purpose AI (GPAI) models. With the exponential growth in size, capability, reach and accompanying risk of these models, ensuring accountability, safety, and public trust requires frameworks that go beyond traditional black-box methods. The discussion begins with an analysis of the need for deeper-than-black-box evaluations (Section I), emphasizing the importance of understanding model internals to uncover latent risks and ensure compliance with ethical and regulatory standards. Building on this foundation, Section II addresses the security considerations of remote evaluations, outlining the threat landscape, technical solutions, and safeguards necessary to protect both evaluators and proprietary model data. Finally, Section III synthesizes these insights into actionable recommendations and future directions, aiming to establish a robust, scalable, and transparent framework for external assessments in GPAI governance.
Open AccessSecuring External Deeper-than-black-box GPAI Evaluations
This paper examines the critical challenges and potential solutions for conducting secure and effective external evaluations of general-purpose AI (GPAI) models. With the exponential growth in size, capability, reach and accompanying risk of these models, ensuring accountability, safety, and public trust requires frameworks that go beyond traditional black-box methods. The discussion begins with an analysis of the need for deeper-than-black-box evaluations (Section I), emphasizing the importance of understanding model internals to uncover latent risks and ensure compliance with ethical and regulatory standards. Building on this foundation, Section II addresses the security considerations of remote evaluations, outlining the threat landscape, technical solutions, and safeguards necessary to protect both evaluators and proprietary model data. Finally, Section III synthesizes these insights into actionable recommendations and future directions, aiming to establish a robust, scalable, and transparent framework for external assessments in GPAI governance.
- Open Access
- 11 March 2025
- Published by eLife Sciences Publications, Ltd in eLife preprints
- Vol. 13
Diverging roles of TRPV1 and TRPM2 in warm-temperature detection
The accurate perception of innocuous temperatures, particularly those experienced as pleasantly warm, is essential for achieving thermal comfort and maintaining thermoregulatory balance. Warm-sensitive neurons (WSN) innervating the skin play a central role in non-painful warmth detection. The TRP ion channels TRPV1 and TRPM2 have been suggested as sensors of warm temperature in WSNs. However, the precise contribution of these channels to the process of warmth detection is not fully understood. A significant challenge in analysing WSNs lies in their scarcity: fewer than 10 % of sensory neurons in the rodent dorsal root ganglion (DRG) respond to innocuous warm temperatures. In this study, we examined >20,000 cultured mouse DRG neurons using calcium imaging and discovered distinct contributions of TRPV1 and TRPM2 to warm-temperature sensitivity. TRPV1, and to a lesser extent TRPM2, affect the abundance of WSNs, with TRPV1 mediating the rapid, dynamic response to warmth and TRPM2 subtly affecting the population response of WSNs. By carefully tracking animal movement in a whole-body thermal preference paradigm, we observe that these cellular differences correlate with nuanced thermal behaviours. Utilizing a drift-diffusion model to quantitatively analyse the decision-making process of animals exposed to different environmental temperatures, we found that: TRPV1 deletion primarily impairs the precision of evidence accumulation, whereas TRPM2 deletion significantly increases the total duration of exposure to warmer environments that are avoided by wildtype mice. Our findings provide valuable insights into the distinct molecular responses to warmth stimuli, and underpin the subtle aspects of thermal decision-making when encountering minor temperature variations.
- 11 March 2025
- Published by eLife Sciences Publications, Ltd in eLife preprints
- Vol. 13
Open AccessDiverging roles of TRPV1 and TRPM2 in warm-temperature detection
The accurate perception of innocuous temperatures, particularly those experienced as pleasantly warm, is essential for achieving thermal comfort and maintaining thermoregulatory balance. Warm-sensitive neurons (WSN) innervating the skin play a central role in non-painful warmth detection. The TRP ion channels TRPV1 and TRPM2 have been suggested as sensors of warm temperature in WSNs. However, the precise contribution of these channels to the process of warmth detection is not fully understood. A significant challenge in analysing WSNs lies in their scarcity: fewer than 10 % of sensory neurons in the rodent dorsal root ganglion (DRG) respond to innocuous warm temperatures. In this study, we examined >20,000 cultured mouse DRG neurons using calcium imaging and discovered distinct contributions of TRPV1 and TRPM2 to warm-temperature sensitivity. TRPV1, and to a lesser extent TRPM2, affect the abundance of WSNs, with TRPV1 mediating the rapid, dynamic response to warmth and TRPM2 subtly affecting the population response of WSNs. By carefully tracking animal movement in a whole-body thermal preference paradigm, we observe that these cellular differences correlate with nuanced thermal behaviours. Utilizing a drift-diffusion model to quantitatively analyse the decision-making process of animals exposed to different environmental temperatures, we found that: TRPV1 deletion primarily impairs the precision of evidence accumulation, whereas TRPM2 deletion significantly increases the total duration of exposure to warmer environments that are avoided by wildtype mice. Our findings provide valuable insights into the distinct molecular responses to warmth stimuli, and underpin the subtle aspects of thermal decision-making when encountering minor temperature variations.
- Open Access
Sharing the Spotlight: Uncovering common attentional dynamics across species
Sustained attention is a key underlying process to many natural behaviours that are shared by multiple species. Yet the way it is commonly studied in a lab context precludes any meaningful cross-species comparisons. Here, we engaged mice, monkeys, and humans in the exact same, natural perceptual decision task in a VR environment. We captured their behaviour into several parameters along the speed/accuracy axes along which sustained attention is classically defined, and used HMMs to infer four attentional states. We show that the dynamics of these states, both in terms of their durations as well as the transitions between them, are much more similar between the species than might have been expected. Moreover, attentional state fluctuations seem to be internally generated. The task and analyses developed here represent a new approach to comparing sustained attention across species in an objective, data-driven way.
Open AccessSharing the Spotlight: Uncovering common attentional dynamics across species
Sustained attention is a key underlying process to many natural behaviours that are shared by multiple species. Yet the way it is commonly studied in a lab context precludes any meaningful cross-species comparisons. Here, we engaged mice, monkeys, and humans in the exact same, natural perceptual decision task in a VR environment. We captured their behaviour into several parameters along the speed/accuracy axes along which sustained attention is classically defined, and used HMMs to infer four attentional states. We show that the dynamics of these states, both in terms of their durations as well as the transitions between them, are much more similar between the species than might have been expected. Moreover, attentional state fluctuations seem to be internally generated. The task and analyses developed here represent a new approach to comparing sustained attention across species in an objective, data-driven way.
- Open Access
- 19 April 2024
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 20 (4) , e1012000
What does the mean mean? A simple test for neuroscience
Trial-averaged metrics, e.g. tuning curves or population response vectors, are a ubiquitous way of characterizing neuronal activity. But how relevant are such trial-averaged responses to neuronal computation itself? Here we present a simple test to estimate whether average responses reflect aspects of neuronal activity that contribute to neuronal processing. The test probes two assumptions implicitly made whenever average metrics are treated as meaningful representations of neuronal activity: Reliability: Neuronal responses repeat consistently enough across trials that they convey a recognizable reflection of the average response to downstream regions. Behavioural relevance: If a single-trial response is more similar to the average template, it is more likely to evoke correct behavioural responses. We apply this test to two data sets: (1) Two-photon recordings in primary somatosensory cortices (S1 and S2) of mice trained to detect optogenetic stimulation in S1; and (2) Electrophysiological recordings from 71 brain areas in mice performing a contrast discrimination task. Under the highly controlled settings of Data set 1, both assumptions were largely fulfilled. In contrast, the less restrictive paradigm of Data set 2 met neither assumption. Simulations predict that the larger diversity of neuronal response preferences, rather than higher cross-trial reliability, drives the better performance of Data set 1. We conclude that when behaviour is less tightly restricted, average responses do not seem particularly relevant to neuronal computation, potentially because information is encoded more dynamically. Most importantly, we encourage researchers to apply this simple test of computational relevance whenever using trial-averaged neuronal metrics, in order to gauge how representative cross-trial averages are in a given context.
- 19 April 2024
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 20 (4) , e1012000
Open AccessWhat does the mean mean? A simple test for neuroscience
Trial-averaged metrics, e.g. tuning curves or population response vectors, are a ubiquitous way of characterizing neuronal activity. But how relevant are such trial-averaged responses to neuronal computation itself? Here we present a simple test to estimate whether average responses reflect aspects of neuronal activity that contribute to neuronal processing. The test probes two assumptions implicitly made whenever average metrics are treated as meaningful representations of neuronal activity: Reliability: Neuronal responses repeat consistently enough across trials that they convey a recognizable reflection of the average response to downstream regions. Behavioural relevance: If a single-trial response is more similar to the average template, it is more likely to evoke correct behavioural responses. We apply this test to two data sets: (1) Two-photon recordings in primary somatosensory cortices (S1 and S2) of mice trained to detect optogenetic stimulation in S1; and (2) Electrophysiological recordings from 71 brain areas in mice performing a contrast discrimination task. Under the highly controlled settings of Data set 1, both assumptions were largely fulfilled. In contrast, the less restrictive paradigm of Data set 2 met neither assumption. Simulations predict that the larger diversity of neuronal response preferences, rather than higher cross-trial reliability, drives the better performance of Data set 1. We conclude that when behaviour is less tightly restricted, average responses do not seem particularly relevant to neuronal computation, potentially because information is encoded more dynamically. Most importantly, we encourage researchers to apply this simple test of computational relevance whenever using trial-averaged neuronal metrics, in order to gauge how representative cross-trial averages are in a given context.
- Open Access
- 29 March 2024
- Published by Springer Nature in Research Square
Thoughtful faces: inferring internal states across species using facial features
Animal behaviour is shaped to a large degree by internal cognitive states, but it is unknown whether these states are similar across species. To address this question, we developed a virtual reality setup in which mice and macaques engage in the same naturalistic visual foraging task. We exploited the richness of a wide range of facial features extracted from video recordings during the task, to train a Markov-Switching Linear Regression (MSLR). By doing so, we identified, on a single-trial basis, a set of internal states that reliably predicted when the animals were going to react to the presented stimuli. Even though the model was trained purely on reaction times, it could also predict task outcome, supporting the behavioural relevance of the inferred states. The identified states were comparable between mice and monkeys. Furthermore, each state corresponded to a characteristic pattern of facial features, highlighting the importance of facial expressions as manifestations of internal cognitive states across species.
- 29 March 2024
- Published by Springer Nature in Research Square
Open AccessThoughtful faces: inferring internal states across species using facial features
Animal behaviour is shaped to a large degree by internal cognitive states, but it is unknown whether these states are similar across species. To address this question, we developed a virtual reality setup in which mice and macaques engage in the same naturalistic visual foraging task. We exploited the richness of a wide range of facial features extracted from video recordings during the task, to train a Markov-Switching Linear Regression (MSLR). By doing so, we identified, on a single-trial basis, a set of internal states that reliably predicted when the animals were going to react to the presented stimuli. Even though the model was trained purely on reaction times, it could also predict task outcome, supporting the behavioural relevance of the inferred states. The identified states were comparable between mice and monkeys. Furthermore, each state corresponded to a characteristic pattern of facial features, highlighting the importance of facial expressions as manifestations of internal cognitive states across species.
- Open Access
Inferring internal states across mice and monkeys using facial features
Animal behaviour is shaped to a large degree by internal cognitive states, but it is unknown whether these states are similar across species. To address this question, here we develop a virtual reality setup in which male mice and macaques engage in the same naturalistic visual foraging task. We exploit the richness of a wide range of facial features extracted from video recordings during the task, to train a Markov-Switching Linear Regression (MSLR). By doing so, we identify, on a singletrial basis, a set of internal states that reliably predicts when the animals are going to react to the presented stimuli. Even though the model is trained purely on reaction times, it can also predict task outcome, supporting the behavioural relevance of the inferred states. The relationship of the identified states to task performance is comparable between mice and monkeys. Furthermore, each state corresponds to a characteristic pattern of facial features that partially overlaps between species, highlighting the importance of facial expressions as manifestations of internal cognitive states across species.
Open AccessInferring internal states across mice and monkeys using facial features
Animal behaviour is shaped to a large degree by internal cognitive states, but it is unknown whether these states are similar across species. To address this question, here we develop a virtual reality setup in which male mice and macaques engage in the same naturalistic visual foraging task. We exploit the richness of a wide range of facial features extracted from video recordings during the task, to train a Markov-Switching Linear Regression (MSLR). By doing so, we identify, on a singletrial basis, a set of internal states that reliably predicts when the animals are going to react to the presented stimuli. Even though the model is trained purely on reaction times, it can also predict task outcome, supporting the behavioural relevance of the inferred states. The relationship of the identified states to task performance is comparable between mice and monkeys. Furthermore, each state corresponds to a characteristic pattern of facial features that partially overlaps between species, highlighting the importance of facial expressions as manifestations of internal cognitive states across species.
- Open Access
Diverging roles of TRPV1 and TRPM2 in warm-temperature detection
The accurate perception of innocuous temperatures, particularly those experienced as pleasantly warm, is essential for achieving thermal comfort and maintaining thermoregulatory balance. Warm-sensitive neurons (WSN) innervating the skin play a central role in non-painful warmth detection. The TRP ion channels TRPV1 and TRPM2 have been suggested as sensors of warm temperature in WSNs. However, the precise contribution of these channels to the process of warmth detection is not fully understood. A significant challenge in analysing WSNs lies in their scarcity: fewer than 10 % of sensory neurons in the rodent dorsal root ganglion (DRG) respond to innocuous warm temperatures. In this study, we examined >20,000 cultured mouse DRG neurons using calcium imaging and discovered distinct contributions of TRPV1 and TRPM2 to warm-temperature sensitivity. TRPV1, and to a lesser extent TRPM2, affect the abundance of WSNs, with TRPV1 mediating the rapid, dynamic response to warmth and TRPM2 subtly affecting the population response of WSNs. By carefully tracking animal movement in a whole-body thermal preference paradigm, we observe that these cellular differences correlate with nuanced thermal behaviours. Utilizing a drift-diffusion model to quantitatively analyse the decision-making process of animals exposed to different environmental temperatures, we found that: TRPV1 deletion primarily impairs the precision of evidence accumulation, whereas TRPM2 deletion significantly increases the total duration of exposure to warmer environments that are avoided by wildtype mice. Our findings provide valuable insights into the distinct molecular responses to warmth stimuli, and underpin the subtle aspects of thermal decision-making when encountering minor temperature variations.
Open AccessDiverging roles of TRPV1 and TRPM2 in warm-temperature detection
The accurate perception of innocuous temperatures, particularly those experienced as pleasantly warm, is essential for achieving thermal comfort and maintaining thermoregulatory balance. Warm-sensitive neurons (WSN) innervating the skin play a central role in non-painful warmth detection. The TRP ion channels TRPV1 and TRPM2 have been suggested as sensors of warm temperature in WSNs. However, the precise contribution of these channels to the process of warmth detection is not fully understood. A significant challenge in analysing WSNs lies in their scarcity: fewer than 10 % of sensory neurons in the rodent dorsal root ganglion (DRG) respond to innocuous warm temperatures. In this study, we examined >20,000 cultured mouse DRG neurons using calcium imaging and discovered distinct contributions of TRPV1 and TRPM2 to warm-temperature sensitivity. TRPV1, and to a lesser extent TRPM2, affect the abundance of WSNs, with TRPV1 mediating the rapid, dynamic response to warmth and TRPM2 subtly affecting the population response of WSNs. By carefully tracking animal movement in a whole-body thermal preference paradigm, we observe that these cellular differences correlate with nuanced thermal behaviours. Utilizing a drift-diffusion model to quantitatively analyse the decision-making process of animals exposed to different environmental temperatures, we found that: TRPV1 deletion primarily impairs the precision of evidence accumulation, whereas TRPM2 deletion significantly increases the total duration of exposure to warmer environments that are avoided by wildtype mice. Our findings provide valuable insights into the distinct molecular responses to warmth stimuli, and underpin the subtle aspects of thermal decision-making when encountering minor temperature variations.