{"id":{"repo_id":"usf","oai_identifier":"oai:digitalcommons.usf.edu:etd-11474"},"canonical_url":"https://search.ndltd.org/etd/usf/oai:digitalcommons.usf.edu:etd-11474","repository":{"repo_id":"usf","name":"University of South Florida","base_url":"https://digitalcommons.usf.edu/do/oai/"},"display":{"title":"Individual Behavioral Modeling Across Games of Strategy","abstract":"An individual’s actions in a particular environment and with specified resources can reveal their decision-making tendencies and patterns, and by analyzing the variations in cognitive traits among individuals, it may be possible to identify trends that can foretell their future behaviors. This can be a powerful tool in various fields including cognitive modeling, player analytics, computer security, and threat detection. Collectible card games are a fruitful test space for studying cognitive differences in decision-making, as they can have clearly defined and replicable environments and large player bases. As such, in this work, I explore the potential of using two virtual collectible card games, Legends of Code and Magic and Hearthstone, to distinguish individual players based on their gameplay and predict the future actions of a player whose behavior is already known. However, the task of defining and fixing environments within a game is nontrivial. Additionally, it would be naive to assume we could reliably obtain enough behavioral information about an individual from any singular strategic game to make decisive predictions about their cognitive traits. As such, this work also explores the ability to identify cognitive traits across multiple strategic tasks, using the commonly studied Iowa Gambling Task and Blackjack. The results indicate that, within strategic games, the maximum possible environmental information is necessary to reach full capability in identifying players and predicting their future actions. Additionally, making predictions about a player’s future behavior by using individualized models is much more accurate than relying on generalized data from many players. Finally, some cognitive traits, namely confidence and working memory, may be transferable between the Iowa Gambling Task and Blackjack.","abstract_html":"An individual’s actions in a particular environment and with specified resources can reveal their decision-making tendencies and patterns, and by analyzing the variations in cognitive traits among individuals, it may be possible to identify trends that can foretell their future behaviors. This can be a powerful tool in various fields including cognitive modeling, player analytics, computer security, and threat detection. Collectible card games are a fruitful test space for studying cognitive differences in decision-making, as they can have clearly defined and replicable environments and large player bases. As such, in this work, I explore the potential of using two virtual collectible card games, Legends of Code and Magic and Hearthstone, to distinguish individual players based on their gameplay and predict the future actions of a player whose behavior is already known. However, the task of defining and fixing environments within a game is nontrivial. Additionally, it would be naive to assume we could reliably obtain enough behavioral information about an individual from any singular strategic game to make decisive predictions about their cognitive traits. As such, this work also explores the ability to identify cognitive traits across multiple strategic tasks, using the commonly studied Iowa Gambling Task and Blackjack. The results indicate that, within strategic games, the maximum possible environmental information is necessary to reach full capability in identifying players and predicting their future actions. Additionally, making predictions about a player’s future behavior by using individualized models is much more accurate than relying on generalized data from many players. Finally, some cognitive traits, namely confidence and working memory, may be transferable between the Iowa Gambling Task and Blackjack.","abstract_has_math":false,"creators":["Fields, Logan"],"institution":"University of South Florida","degree_name":"Doctor of Philosophy (Ph.D.)","degree_level":"Dissertation","degree_discipline":"Computer Science and Engineering","degree_department":null,"school":null,"contributors":["John Licato, Ph.D.","Michael Gillespie, Ph.D.","Seungbae Kim, Ph.D.","Susana Lai-Yuen, Ph.D.","Ankur Mali, Ph.D."],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2024,"date_issued":"2024-03-29T07:00:00Z","date_published":"2024-03-29T07:00:00Z","updated_at":"2026-11-19T08:48:51Z","subjects":["Blackjack","Collectible card games","Iowa Gambling Task","Language models","NLP","Player modeling","Behavioral Disciplines and Activities","Computer Sciences"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://digitalcommons.usf.edu/etd/10187","outbound_label":"Repository record","outbound_source":"dc:identifier","record_url":"https://digitalcommons.usf.edu/etd/10187","record_label":"Repository record","record_kind":"record","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://digitalcommons.usf.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Adigitalcommons.usf.edu%3Aetd-11474","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["John Licato, Ph.D.","Michael Gillespie, Ph.D.","Seungbae Kim, Ph.D.","Susana Lai-Yuen, Ph.D.","Ankur Mali, Ph.D."]},{"key":"dc:creator","label":"Author","values":["Fields, Logan"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2024-03-29T07:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science and Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Dissertation","Ph.D."]},{"key":"thesis:degree_name","label":"Degree Name","values":["Doctor of Philosophy (Ph.D.)"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of South Florida"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Blackjack","Collectible card games","Iowa Gambling Task","Language models","NLP","Player modeling","Behavioral Disciplines and Activities","Computer Sciences"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.usf.edu/etd/10187"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["An individual’s actions in a particular environment and with specified resources can reveal their decision-making tendencies and patterns, and by analyzing the variations in cognitive traits among individuals, it may be possible to identify trends that can foretell their future behaviors. This can be a powerful tool in various fields including cognitive modeling, player analytics, computer security, and threat detection. Collectible card games are a fruitful test space for studying cognitive differences in decision-making, as they can have clearly defined and replicable environments and large player bases. As such, in this work, I explore the potential of using two virtual collectible card games, Legends of Code and Magic and Hearthstone, to distinguish individual players based on their gameplay and predict the future actions of a player whose behavior is already known. However, the task of defining and fixing environments within a game is nontrivial. Additionally, it would be naive to assume we could reliably obtain enough behavioral information about an individual from any singular strategic game to make decisive predictions about their cognitive traits. As such, this work also explores the ability to identify cognitive traits across multiple strategic tasks, using the commonly studied Iowa Gambling Task and Blackjack. The results indicate that, within strategic games, the maximum possible environmental information is necessary to reach full capability in identifying players and predicting their future actions. Additionally, making predictions about a player’s future behavior by using individualized models is much more accurate than relying on generalized data from many players. Finally, some cognitive traits, namely confidence and working memory, may be transferable between the Iowa Gambling Task and Blackjack."]},{"key":"dc:title","label":"Title","values":["Individual Behavioral Modeling Across Games of Strategy"]}]}],"canonical_facts":{"dc:contributor":["John Licato, Ph.D.","Michael Gillespie, Ph.D.","Seungbae Kim, Ph.D.","Susana Lai-Yuen, Ph.D.","Ankur Mali, Ph.D."],"dc:creator":["Fields, Logan"],"dc:date":["2024-03-29T07:00:00Z"],"dc:description":["An individual’s actions in a particular environment and with specified resources can reveal their decision-making tendencies and patterns, and by analyzing the variations in cognitive traits among individuals, it may be possible to identify trends that can foretell their future behaviors. This can be a powerful tool in various fields including cognitive modeling, player analytics, computer security, and threat detection. Collectible card games are a fruitful test space for studying cognitive differences in decision-making, as they can have clearly defined and replicable environments and large player bases. As such, in this work, I explore the potential of using two virtual collectible card games, Legends of Code and Magic and Hearthstone, to distinguish individual players based on their gameplay and predict the future actions of a player whose behavior is already known. However, the task of defining and fixing environments within a game is nontrivial. Additionally, it would be naive to assume we could reliably obtain enough behavioral information about an individual from any singular strategic game to make decisive predictions about their cognitive traits. As such, this work also explores the ability to identify cognitive traits across multiple strategic tasks, using the commonly studied Iowa Gambling Task and Blackjack. The results indicate that, within strategic games, the maximum possible environmental information is necessary to reach full capability in identifying players and predicting their future actions. Additionally, making predictions about a player’s future behavior by using individualized models is much more accurate than relying on generalized data from many players. Finally, some cognitive traits, namely confidence and working memory, may be transferable between the Iowa Gambling Task and Blackjack."],"dc:identifier":["https://digitalcommons.usf.edu/etd/10187"],"dc:subject":["Blackjack","Collectible card games","Iowa Gambling Task","Language models","NLP","Player modeling","Behavioral Disciplines and Activities","Computer Sciences"],"dc:title":["Individual Behavioral Modeling Across Games of Strategy"],"thesis:degree_discipline":["Computer Science and Engineering"],"thesis:degree_level":["Dissertation","Ph.D."],"thesis:degree_name":["Doctor of Philosophy (Ph.D.)"],"thesis:institution_name":["University of South Florida"]},"updated_at":"2026-11-19T08:48:51Z"}