Pub Date : 1900-01-01DOI: 10.32470/ccn.2019.1340-0
Ciara A Devine, David P. McGovern, Jessica Dully, Emmet McNickle, S. Kelly, R. O’Connell
In perceptual decision-making, dynamic urgency is a time-dependent, evidence-independent mechanism that imposes a gradual reduction in the amount of sensory evidence required to commit to a choice. Although the effects of urgency have been observed across the sensorimotor hierarchy during perceptual decision formation, a distinct neural signature of urgency has yet to be fully characterised in the human brain. Here we tested the hypothesis that the contingent negative variation (CNV), a frontocentral, negative-going potential that has been implicated in temporal processing, directly represents dynamic urgency in the human brain. To this end we analysed data from two experiments in which speed emphasis was manipulated while subjects performed perceptual discrimination tasks. We found that the CNV was more pronounced at baseline under speed pressure, reflecting a static urgency component and that it became more pronounced over time, reflecting a dynamic component. Moreover, we also found that the rate of build up of the CNV accelerated as time elapsed and was not driven by sensory evidence accumulation. Together these findings support the mechanistic characterisation of the CNV as a timedependent, evidence independent urgency signal.
{"title":"An Electrophysiological Signature of Dynamic Urgency in Human Perceptual Decision Making","authors":"Ciara A Devine, David P. McGovern, Jessica Dully, Emmet McNickle, S. Kelly, R. O’Connell","doi":"10.32470/ccn.2019.1340-0","DOIUrl":"https://doi.org/10.32470/ccn.2019.1340-0","url":null,"abstract":"In perceptual decision-making, dynamic urgency is a time-dependent, evidence-independent mechanism that imposes a gradual reduction in the amount of sensory evidence required to commit to a choice. Although the effects of urgency have been observed across the sensorimotor hierarchy during perceptual decision formation, a distinct neural signature of urgency has yet to be fully characterised in the human brain. Here we tested the hypothesis that the contingent negative variation (CNV), a frontocentral, negative-going potential that has been implicated in temporal processing, directly represents dynamic urgency in the human brain. To this end we analysed data from two experiments in which speed emphasis was manipulated while subjects performed perceptual discrimination tasks. We found that the CNV was more pronounced at baseline under speed pressure, reflecting a static urgency component and that it became more pronounced over time, reflecting a dynamic component. Moreover, we also found that the rate of build up of the CNV accelerated as time elapsed and was not driven by sensory evidence accumulation. Together these findings support the mechanistic characterisation of the CNV as a timedependent, evidence independent urgency signal.","PeriodicalId":281121,"journal":{"name":"2019 Conference on Cognitive Computational Neuroscience","volume":"122 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128137457","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.32470/ccn.2019.1117-0
A. Bastos, Omar Costilla-Reyes, E. Miller
Cerebral cortex is composed of 6 anatomical layers. How these layers contribute to computations that give rise to cognition remains a challenge in neuroscience. Part of this challenge is to reliably identify laminar markers from in-vivo neurophysiological data. Classic methods for laminar identification are based on assumptions which are often violated and require expert users to identify the pattern, potentially introducing bias. We recorded local field potentials (LFP) with probes containing 16 or 32 electrodes that span all cortical layers in frontal, parietal, and visual cortex in monkeys. We describe two novel methods to identify layers in a fully automatic and quantitative way. The first method represents relative power across electrodes from as a 2-dimensional image, and maximizes image similarity across probes. The second method leverages ensemble machine learning to maximize classification accuracy of LFP data to a laminar label. Both methods detect consistent patterns, and the image similarity approach reveals a cortex-wide motif of laminar expression for delta/theta, alpha/beta and gamma rhythms. Delta/theta (1-4 Hz) and gamma (50150 Hz) power peak in superficial layers 2/3, and alpha/beta (10-30 Hz) power peaks in deep layers 5/6.
{"title":"Automatic methods for cortex-wide layer identification of electrophysiological signals reveals a cortical motif for the expression of neuronal rhythms","authors":"A. Bastos, Omar Costilla-Reyes, E. Miller","doi":"10.32470/ccn.2019.1117-0","DOIUrl":"https://doi.org/10.32470/ccn.2019.1117-0","url":null,"abstract":"Cerebral cortex is composed of 6 anatomical layers. How these layers contribute to computations that give rise to cognition remains a challenge in neuroscience. Part of this challenge is to reliably identify laminar markers from in-vivo neurophysiological data. Classic methods for laminar identification are based on assumptions which are often violated and require expert users to identify the pattern, potentially introducing bias. We recorded local field potentials (LFP) with probes containing 16 or 32 electrodes that span all cortical layers in frontal, parietal, and visual cortex in monkeys. We describe two novel methods to identify layers in a fully automatic and quantitative way. The first method represents relative power across electrodes from as a 2-dimensional image, and maximizes image similarity across probes. The second method leverages ensemble machine learning to maximize classification accuracy of LFP data to a laminar label. Both methods detect consistent patterns, and the image similarity approach reveals a cortex-wide motif of laminar expression for delta/theta, alpha/beta and gamma rhythms. Delta/theta (1-4 Hz) and gamma (50150 Hz) power peak in superficial layers 2/3, and alpha/beta (10-30 Hz) power peaks in deep layers 5/6.","PeriodicalId":281121,"journal":{"name":"2019 Conference on Cognitive Computational Neuroscience","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128093303","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.32470/ccn.2019.1208-0
F. Sandhaeger, Nina Omejc, Anna-Antonia Pape, M. Siegel
Choice Representations Generalize Between Task Contexts Florian Sandhaeger (florian.sandhaeger@uni-tuebingen.de) Hertie-Institute for Clinical Brain Research, Centre for Integrative Neuroscience & MEG Center University of Tuebingen, Otfried-Mueller-Str. 25, 72076 Tuebingen, Germany Nina Omejc (nina.omejc@student.uni-tuebingen.de) Hertie-Institute for Clinical Brain Research, Centre for Integrative Neuroscience & MEG Center, University of Tuebingen, Otfried-Mueller-Str. 25, 72076 Tuebingen, Germany Anna-Antonia Pape (anna.antonia.pape@gmail.com) Centre for Integrative Neuroscience & MEG Center University of Tuebingen, Otfried-Mueller-Str. 25, 72076 Tuebingen, Germany Markus Siegel (markus.siegel@uni-tuebingen.de) Hertie-Institute for Clinical Brain Research, Centre for Integrative Neuroscience & MEG Center, University of Tuebingen, Otfried-Mueller-Str. 25, 72076 Tuebingen, Germany
{"title":"Abstract Choice Representations Generalize Between Task Contexts","authors":"F. Sandhaeger, Nina Omejc, Anna-Antonia Pape, M. Siegel","doi":"10.32470/ccn.2019.1208-0","DOIUrl":"https://doi.org/10.32470/ccn.2019.1208-0","url":null,"abstract":"Choice Representations Generalize Between Task Contexts Florian Sandhaeger (florian.sandhaeger@uni-tuebingen.de) Hertie-Institute for Clinical Brain Research, Centre for Integrative Neuroscience & MEG Center University of Tuebingen, Otfried-Mueller-Str. 25, 72076 Tuebingen, Germany Nina Omejc (nina.omejc@student.uni-tuebingen.de) Hertie-Institute for Clinical Brain Research, Centre for Integrative Neuroscience & MEG Center, University of Tuebingen, Otfried-Mueller-Str. 25, 72076 Tuebingen, Germany Anna-Antonia Pape (anna.antonia.pape@gmail.com) Centre for Integrative Neuroscience & MEG Center University of Tuebingen, Otfried-Mueller-Str. 25, 72076 Tuebingen, Germany Markus Siegel (markus.siegel@uni-tuebingen.de) Hertie-Institute for Clinical Brain Research, Centre for Integrative Neuroscience & MEG Center, University of Tuebingen, Otfried-Mueller-Str. 25, 72076 Tuebingen, Germany","PeriodicalId":281121,"journal":{"name":"2019 Conference on Cognitive Computational Neuroscience","volume":"156 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130460080","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.32470/ccn.2019.1326-0
Katrina R. Quinn, B. Cumming, H. Nienborg
Many V2 neurons are selective for binocular disparity. V2 is also the earliest site in the visual processing hierarchy for which systematic correlations across the population between neural responses and an animal’s behavioral choice in disparity based tasks have been observed. However, while these choice correlations suggest a link between the neural activity and perceptual choice, it has long been recognized that they do not establish a causal relationship. Here, we sought to test whether macaque V2 plays a causal role on coarse disparity judgements. We used microstimulation on disparity selective sites in V2 whilst animals performed a coarse disparity discrimination task. We found that microstimulation led to a systematic shift of the psychometric function towards the preferred disparity of the stimulated site, supporting a causal role for V2 neurons in disparity discrimination.
{"title":"A Causal Effect of Macaque V2 in a Coarse Disparity Discrimination Task","authors":"Katrina R. Quinn, B. Cumming, H. Nienborg","doi":"10.32470/ccn.2019.1326-0","DOIUrl":"https://doi.org/10.32470/ccn.2019.1326-0","url":null,"abstract":"Many V2 neurons are selective for binocular disparity. V2 is also the earliest site in the visual processing hierarchy for which systematic correlations across the population between neural responses and an animal’s behavioral choice in disparity based tasks have been observed. However, while these choice correlations suggest a link between the neural activity and perceptual choice, it has long been recognized that they do not establish a causal relationship. Here, we sought to test whether macaque V2 plays a causal role on coarse disparity judgements. We used microstimulation on disparity selective sites in V2 whilst animals performed a coarse disparity discrimination task. We found that microstimulation led to a systematic shift of the psychometric function towards the preferred disparity of the stimulated site, supporting a causal role for V2 neurons in disparity discrimination.","PeriodicalId":281121,"journal":{"name":"2019 Conference on Cognitive Computational Neuroscience","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130092465","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.32470/ccn.2019.1416-0
Tyler Brooke-Wilson
A crucial methodological question for cognitive neuroscience is the question of what constitutes evidence of neural representation. A number of critiques over the last decade have challenged the view that correlation alone, as measured by neural decoding, is sufficient to establish representation. In response to such critiques, correlation is often augmented by a behavioral measure, showing that the decoding accuracy of a classifier and some behavioral performance measure are themselves correlated. I argue that correlation and behavioral causation together are nevertheless still insufficient for establishing representation. Inferring the existence of a neural representation on the basis of correlation and behavior alone is liable to both false positives and false negatives. Reflection on one common theory of representation (functional homomorphism theory, proposed by King and Gallistel 2010) elucidates why correlation + behavior is insufficient and suggests more direct sources of evidence. I present this theory and explain its implications for the question of empirical evidence of representation. Along the way I draw out some of the connections between the functional homomorphism theory of representation and predictive theories of perception.
{"title":"Sources of Evidence for Neural Representation","authors":"Tyler Brooke-Wilson","doi":"10.32470/ccn.2019.1416-0","DOIUrl":"https://doi.org/10.32470/ccn.2019.1416-0","url":null,"abstract":"A crucial methodological question for cognitive neuroscience is the question of what constitutes evidence of neural representation. A number of critiques over the last decade have challenged the view that correlation alone, as measured by neural decoding, is sufficient to establish representation. In response to such critiques, correlation is often augmented by a behavioral measure, showing that the decoding accuracy of a classifier and some behavioral performance measure are themselves correlated. I argue that correlation and behavioral causation together are nevertheless still insufficient for establishing representation. Inferring the existence of a neural representation on the basis of correlation and behavior alone is liable to both false positives and false negatives. Reflection on one common theory of representation (functional homomorphism theory, proposed by King and Gallistel 2010) elucidates why correlation + behavior is insufficient and suggests more direct sources of evidence. I present this theory and explain its implications for the question of empirical evidence of representation. Along the way I draw out some of the connections between the functional homomorphism theory of representation and predictive theories of perception.","PeriodicalId":281121,"journal":{"name":"2019 Conference on Cognitive Computational Neuroscience","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129882902","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.32470/ccn.2019.1169-0
Krista Bond, Alexis Porter, T. Verstynen
Humans and other mammals flexibly select actions in noisy, uncertain contexts, quickly using feedback to adapt their decision policies to either explore other options or to exploit what they know. Drawing inspiration from the plasticity of cortico-basal ganglia-thalamic circuitry, we recently developed a cognitive model of decision-making that uses both a value-driven learning signal to update an internal estimate of state action-value (i.e., conflict in the probability of reward between two choices) and a change-point-driven learning signal that adapts to changes in reward contingencies (i.e., a previously high value target becoming devalued). In this work, we expand on previous results from our group (Bond, Dunovan, & Verstynen, 2018) to more carefully detail how these environmental signals drive changes in the decision process. Across nine separate behavioral testing sessions, we independently manipulated the level of value-conflict and volatility in action-outcome contingencies. Using a hierarchical drift diffusion model, we found that the belief in the value difference between options had the greatest influence on decision processes, impacting drift rate, while estimates of environmental change had a smaller, but detectable influence on the decision threshold. Taken together, these findings bolster our previous work showing how separate environmental signals impact different aspects of the decision algorithm.
{"title":"A potential reset mechanism for the modulation of decision processes under uncertainty","authors":"Krista Bond, Alexis Porter, T. Verstynen","doi":"10.32470/ccn.2019.1169-0","DOIUrl":"https://doi.org/10.32470/ccn.2019.1169-0","url":null,"abstract":"Humans and other mammals flexibly select actions in noisy, uncertain contexts, quickly using feedback to adapt their decision policies to either explore other options or to exploit what they know. Drawing inspiration from the plasticity of cortico-basal ganglia-thalamic circuitry, we recently developed a cognitive model of decision-making that uses both a value-driven learning signal to update an internal estimate of state action-value (i.e., conflict in the probability of reward between two choices) and a change-point-driven learning signal that adapts to changes in reward contingencies (i.e., a previously high value target becoming devalued). In this work, we expand on previous results from our group (Bond, Dunovan, & Verstynen, 2018) to more carefully detail how these environmental signals drive changes in the decision process. Across nine separate behavioral testing sessions, we independently manipulated the level of value-conflict and volatility in action-outcome contingencies. Using a hierarchical drift diffusion model, we found that the belief in the value difference between options had the greatest influence on decision processes, impacting drift rate, while estimates of environmental change had a smaller, but detectable influence on the decision threshold. Taken together, these findings bolster our previous work showing how separate environmental signals impact different aspects of the decision algorithm.","PeriodicalId":281121,"journal":{"name":"2019 Conference on Cognitive Computational Neuroscience","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130318531","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.32470/ccn.2019.1145-0
Simone Viganò, V. Borghesani, M. Piazza
A fundamental issue in cognitive science is the so-called “symbol-grounding problem” (Harnad 1980), related to the question of how symbols acquire meaning. One simple view posits that, for concrete words, our brain solves the problem by creating associations between the neural representations of the surface forms of symbols (spoken or written words) to the one(s) evoked by the object, action, or event classes the symbols refer to (e.g., see Pulvermuller 2013; 2018). Evidence supporting this view comes from the observation that words related to well known concepts such as numerical quantities (Piazza et al. 2007; Eger et al. 2009), colors (e.g. Simmons et al. 2007), manipulable objects (Chao et al. 1999), places (Kumar et al. 2017), or actions (Hauk 2004; 2011), automatically re-activate the same brain regions that are active during the perception/execution of those specific object features/actions. These data, however, are informative on the neural bases of symbol grounded representations, but not on those underlying symbol grounding: i) they fall short in assessing the role of memory systems implicated in this kind of symbol-toconcept associative learning, and ii) they do not provide a full picture of the effects that symbol grounding has on the brain. Here, to investigate the neural changes generated by this process, we adopted an artificial learning paradigm where 21 adult subjects learned to categorize novel multisensory objects by giving them specific symbolic labels.
认知科学中的一个基本问题是所谓的“符号基础问题”(Harnad 1980)，与符号如何获得意义的问题有关。一种简单的观点认为，对于具体的单词，我们的大脑通过在符号的表面形式(口语或书面文字)的神经表征与符号所指的对象、动作或事件类所唤起的表征之间建立联系来解决问题(例如，参见粉状穆勒2013;2018)。支持这一观点的证据来自于对与众所周知的概念相关的词汇的观察，如数值量(Piazza et al. 2007;Eger等人，2009)，颜色(例如Simmons等人，2007)，可操作对象(Chao等人，1999)，地点(Kumar等人，2017)或动作(Hauk 2004;2011)，自动重新激活在感知/执行这些特定对象特征/动作期间活跃的相同大脑区域。然而，这些数据在符号基础表征的神经基础上提供了信息，但在那些潜在的符号基础上却没有:1)它们在评估涉及这种从符号到概念的联想学习的记忆系统的作用方面存在不足，2)它们没有提供符号基础对大脑的影响的全貌。为了研究这一过程所产生的神经变化，我们采用了人工学习范式，让21名成年受试者通过给予特定的符号标签来学习对新的多感官物体进行分类。
{"title":"How the Human Brain Solves the Symbol-Grounding Problem","authors":"Simone Viganò, V. Borghesani, M. Piazza","doi":"10.32470/ccn.2019.1145-0","DOIUrl":"https://doi.org/10.32470/ccn.2019.1145-0","url":null,"abstract":"A fundamental issue in cognitive science is the so-called “symbol-grounding problem” (Harnad 1980), related to the question of how symbols acquire meaning. One simple view posits that, for concrete words, our brain solves the problem by creating associations between the neural representations of the surface forms of symbols (spoken or written words) to the one(s) evoked by the object, action, or event classes the symbols refer to (e.g., see Pulvermuller 2013; 2018). Evidence supporting this view comes from the observation that words related to well known concepts such as numerical quantities (Piazza et al. 2007; Eger et al. 2009), colors (e.g. Simmons et al. 2007), manipulable objects (Chao et al. 1999), places (Kumar et al. 2017), or actions (Hauk 2004; 2011), automatically re-activate the same brain regions that are active during the perception/execution of those specific object features/actions. These data, however, are informative on the neural bases of symbol grounded representations, but not on those underlying symbol grounding: i) they fall short in assessing the role of memory systems implicated in this kind of symbol-toconcept associative learning, and ii) they do not provide a full picture of the effects that symbol grounding has on the brain. Here, to investigate the neural changes generated by this process, we adopted an artificial learning paradigm where 21 adult subjects learned to categorize novel multisensory objects by giving them specific symbolic labels.","PeriodicalId":281121,"journal":{"name":"2019 Conference on Cognitive Computational Neuroscience","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127946944","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.32470/ccn.2019.1309-0
Benjamin Cuthbert, M. Paré, D. Standage, Gunnar Blohm
Visual working memory experiments typically involve asking a subject to memorize several visual stimuli such as coloured shapes, oriented lines, faces, or objects. Computational accounts of recall performance often assume that each stimulus presented in a trial is encoded independently, ignoring higher-level ensemble statistics that have been shown to bias recall and impact task performance. Here, we analyzed data from a delayed estimation task that required the report of all stimuli (6 coloured squares). We found evidence for serial dependencies in within-trial reports, suggesting that participants clustered similarly coloured stimuli together. These dependencies were supported by estimates of the mutual information of within-trial report distributions. We present a non-parametric clustering model to quantify the clustering properties of randomly-generated stimulus arrays. We believe this is a promising data-driven approach to characterizing the statistical properties of experimental stimuli. Together, these results provide further evidence that humans encode ensemble statistics of visual scenes in working memory.
{"title":"Colour clustering in visual working memory","authors":"Benjamin Cuthbert, M. Paré, D. Standage, Gunnar Blohm","doi":"10.32470/ccn.2019.1309-0","DOIUrl":"https://doi.org/10.32470/ccn.2019.1309-0","url":null,"abstract":"Visual working memory experiments typically involve asking a subject to memorize several visual stimuli such as coloured shapes, oriented lines, faces, or objects. Computational accounts of recall performance often assume that each stimulus presented in a trial is encoded independently, ignoring higher-level ensemble statistics that have been shown to bias recall and impact task performance. Here, we analyzed data from a delayed estimation task that required the report of all stimuli (6 coloured squares). We found evidence for serial dependencies in within-trial reports, suggesting that participants clustered similarly coloured stimuli together. These dependencies were supported by estimates of the mutual information of within-trial report distributions. We present a non-parametric clustering model to quantify the clustering properties of randomly-generated stimulus arrays. We believe this is a promising data-driven approach to characterizing the statistical properties of experimental stimuli. Together, these results provide further evidence that humans encode ensemble statistics of visual scenes in working memory.","PeriodicalId":281121,"journal":{"name":"2019 Conference on Cognitive Computational Neuroscience","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122867121","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.32470/ccn.2019.1098-0
Lynn K. A. Sörensen, Davide Zambrano, H. Slagter, H. Scholte, S. Bohté
Visuo-spatial attention is a key mechanism for selecting goal-relevant information in natural scenes. We here implement a variant of the normalization model of attention into a spiking convolutional neural network, which approximates attentional gain with a change in firing rates. We apply this type of attention with different spatial extents to various levels in the processing hierarchy of a network performing object recognition in natural scenes. We find that close to the average objectsize attentional kernels yield the best performance, equivalent to a rather focused attentional enhancement. Furthermore, manipulating spatial attention within a single level was ineffective as benefits of spatial attention only arose from the combination of early-to-mid level modulations in the network hierarchy. Our results demonstrate that one can efficiently boost performance on the challenging task of recognizing objects in cluttered environments in a large-scale vision model by understanding attentional gain as a more or less precise representation of sensory information.
{"title":"Spatial Attention introduces Behavioral Trade-off in a Large-Scale Spiking Neural Network","authors":"Lynn K. A. Sörensen, Davide Zambrano, H. Slagter, H. Scholte, S. Bohté","doi":"10.32470/ccn.2019.1098-0","DOIUrl":"https://doi.org/10.32470/ccn.2019.1098-0","url":null,"abstract":"Visuo-spatial attention is a key mechanism for selecting goal-relevant information in natural scenes. We here implement a variant of the normalization model of attention into a spiking convolutional neural network, which approximates attentional gain with a change in firing rates. We apply this type of attention with different spatial extents to various levels in the processing hierarchy of a network performing object recognition in natural scenes. We find that close to the average objectsize attentional kernels yield the best performance, equivalent to a rather focused attentional enhancement. Furthermore, manipulating spatial attention within a single level was ineffective as benefits of spatial attention only arose from the combination of early-to-mid level modulations in the network hierarchy. Our results demonstrate that one can efficiently boost performance on the challenging task of recognizing objects in cluttered environments in a large-scale vision model by understanding attentional gain as a more or less precise representation of sensory information.","PeriodicalId":281121,"journal":{"name":"2019 Conference on Cognitive Computational Neuroscience","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126155584","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.32470/ccn.2019.1335-0
M. Speekenbrink
The Kalman filter, combined with heuristic choice rules such as softmax, UCB, and Thompson sampling, has been a popular model to identify the role of uncertainty in exploration in human reinforcement learning. Here we show that the Kalman filter combined with a softmax or UCB choice rule is not fully identifiable. By this structural identifiability, we mean that with unlimited data, the true parameter values are determinable. Perhaps surprisingly, the Kalman filter with Thompson sampling is fully identifiable.
{"title":"Identifiability of Gaussian Bayesian bandit models","authors":"M. Speekenbrink","doi":"10.32470/ccn.2019.1335-0","DOIUrl":"https://doi.org/10.32470/ccn.2019.1335-0","url":null,"abstract":"The Kalman filter, combined with heuristic choice rules such as softmax, UCB, and Thompson sampling, has been a popular model to identify the role of uncertainty in exploration in human reinforcement learning. Here we show that the Kalman filter combined with a softmax or UCB choice rule is not fully identifiable. By this structural identifiability, we mean that with unlimited data, the true parameter values are determinable. Perhaps surprisingly, the Kalman filter with Thompson sampling is fully identifiable.","PeriodicalId":281121,"journal":{"name":"2019 Conference on Cognitive Computational Neuroscience","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126343257","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：