{"id":{"repo_id":"bu","oai_identifier":"oai:open.bu.edu:2144/53698"},"canonical_url":"https://search.ndltd.org/etd/bu/oai:open.bu.edu:2144/53698","repository":{"repo_id":"bu","name":"Boston University","base_url":"https://open.bu.edu/oai/request"},"display":{"title":"Elucidating biobehavioral risk factors for anxiety in early childhood","abstract":"Anxiety disorders are among the most common psychiatric conditions in childhood. Although behavioral inhibition (BI) is a robust predictor of anxiety risk, gold-standard observational assessments are time-intensive and difficult to scale, limiting their utility for early identification. Moreover, other important risk factors span many domains, including child characteristics, parent factors, dyadic relationships, and broader contexts (e.g., family conflict, household income), yet these are often studied in isolation. As a result, it remains unclear which risk markers contribute most strongly to children’s anxiety when considered together. To address this gap, the current dissertation integrates observational and physiological risk markers alongside commonly used parent-report measures from a large and diverse sample of preschool-aged children (n = 250) to predict parent-reported child anxiety, a widely used and accessible indicator that guides parents’ decisions to seek treatment. Study 1 assessed the validity of a novel computer vision approach for assessing children’s behavioral risk for anxiety. A computer vision and machine learning prediction model (FEAR-CAM) was trained to assess children’s fear behaviors, an index of BI, during a modified stranger approach task. Validation analyses indicate strong model performance using both training-testing splits and cross-validation, with accuracy comparable to that of gold-standard behavioral coding methods. Study 2 examined whether and how physiological regulation informs reported anxiety using indicators of children’s moment-by-moment parasympathetic regulation (vagal flexibility) across the stranger approach task. Findings indicate variability in non-linear regulatory patterns, and associations with reported anxiety appear sensitive to analytic specification. Alternative approaches to modeling these links are explored. Study 3 examined 63 risk factors indexing multiple domains of childhood anxiety risk, including biobehavioral measures. Predictors were ranked using SHAP values derived from a trained XGBoost model to examine their contribution to model predictions, with results suggesting that child-level factors, particularly emotion symptoms, BI, and parasympathetic regulation, were the most influential contributors to parent-reported anxiety. Together, these findings highlight the potential of scalable behavioral and physiological assessment approaches, integrated with parent-reported indicators, to improve early identification of children at risk for anxiety.","abstract_html":"Anxiety disorders are among the most common psychiatric conditions in childhood. Although behavioral inhibition (BI) is a robust predictor of anxiety risk, gold-standard observational assessments are time-intensive and difficult to scale, limiting their utility for early identification. Moreover, other important risk factors span many domains, including child characteristics, parent factors, dyadic relationships, and broader contexts (e.g., family conflict, household income), yet these are often studied in isolation. As a result, it remains unclear which risk markers contribute most strongly to children’s anxiety when considered together. To address this gap, the current dissertation integrates observational and physiological risk markers alongside commonly used parent-report measures from a large and diverse sample of preschool-aged children (n = 250) to predict parent-reported child anxiety, a widely used and accessible indicator that guides parents’ decisions to seek treatment. Study 1 assessed the validity of a novel computer vision approach for assessing children’s behavioral risk for anxiety. A computer vision and machine learning prediction model (FEAR-CAM) was trained to assess children’s fear behaviors, an index of BI, during a modified stranger approach task. Validation analyses indicate strong model performance using both training-testing splits and cross-validation, with accuracy comparable to that of gold-standard behavioral coding methods. Study 2 examined whether and how physiological regulation informs reported anxiety using indicators of children’s moment-by-moment parasympathetic regulation (vagal flexibility) across the stranger approach task. Findings indicate variability in non-linear regulatory patterns, and associations with reported anxiety appear sensitive to analytic specification. Alternative approaches to modeling these links are explored. Study 3 examined 63 risk factors indexing multiple domains of childhood anxiety risk, including biobehavioral measures. Predictors were ranked using SHAP values derived from a trained XGBoost model to examine their contribution to model predictions, with results suggesting that child-level factors, particularly emotion symptoms, BI, and parasympathetic regulation, were the most influential contributors to parent-reported anxiety. Together, these findings highlight the potential of scalable behavioral and physiological assessment approaches, integrated with parent-reported indicators, to improve early identification of children at risk for anxiety.","abstract_has_math":false,"creators":["Sem, Kathy"],"institution":"Boston University","degree_name":"Doctor of Philosophy","degree_level":"doctoral","degree_discipline":"Psychological & Brain Sciences","degree_department":null,"school":null,"contributors":[],"advisors":["Wagner, Nicholas"],"committee_chairs":[],"committee_members":[],"year":2026,"date_issued":"2026","date_published":"2026","updated_at":"2026-11-18T19:43:36Z","subjects":["Mental health","Physiological psychology","Behavioral psychology","Anxiety","Biobehavioral risk","Child anxiety risk","Computer vision","Machine learning","Physiological risk"],"languages":["en_US"],"rights":["Attribution-NonCommercial-NoDerivatives 4.0 International"],"rights_urls":["http://creativecommons.org/licenses/by-nc-nd/4.0/"],"identifier_entries":[],"person_identifier_entries":[{"key":"dc:identifier.orcid","label":"ORCID","values":["0000-0002-5248-2561"],"render_values":[{"text":"0000-0002-5248-2561","href":"https://orcid.org/0000-0002-5248-2561","code":true}]}]},"links":{"outbound_url":"https://hdl.handle.net/2144/53698","outbound_label":"Handle","outbound_source":"dc:identifier.uri","record_url":"https://hdl.handle.net/2144/53698","record_label":"Repository record","record_kind":"handle","record_source":"dc:identifier.uri","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://open.bu.edu/oai/request?verb=GetRecord&metadataPrefix=dim&identifier=oai%3Aopen.bu.edu%3A2144%2F53698","prefix":"dim"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Wagner, Nicholas"]},{"key":"dc:creator","label":"Author","values":["Sem, Kathy"]},{"key":"dc:identifier.orcid","label":"ORCID","values":["0000-0002-5248-2561"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2026-10-15T17:37:23Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2026-10-15T17:37:23Z"]},{"key":"dc:date.issued","label":"Date","values":["2026"]},{"key":"dc:date.updated","label":"Dc Date Updated","values":["2026-09-09T16:24:37Z"]},{"key":"dc:type","label":"Dc Type","values":["Thesis/Dissertation"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Psychological & Brain Sciences"]},{"key":"thesis:degree_level","label":"Degree Level","values":["doctoral"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Doctor of Philosophy"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Boston University"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Mental health","Physiological psychology","Behavioral psychology","Anxiety","Biobehavioral risk","Child anxiety risk","Computer vision","Machine learning","Physiological risk"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en_US"]},{"key":"dc:rights","label":"Dc Rights","values":["Attribution-NonCommercial-NoDerivatives 4.0 International"]},{"key":"dc:rights.uri","label":"Rights URI","values":["http://creativecommons.org/licenses/by-nc-nd/4.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/2144/53698"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["2026"]},{"key":"dc:description.abstract","label":"Abstract","values":["Anxiety disorders are among the most common psychiatric conditions in childhood. Although behavioral inhibition (BI) is a robust predictor of anxiety risk, gold-standard observational assessments are time-intensive and difficult to scale, limiting their utility for early identification. Moreover, other important risk factors span many domains, including child characteristics, parent factors, dyadic relationships, and broader contexts (e.g., family conflict, household income), yet these are often studied in isolation. As a result, it remains unclear which risk markers contribute most strongly to children’s anxiety when considered together. To address this gap, the current dissertation integrates observational and physiological risk markers alongside commonly used parent-report measures from a large and diverse sample of preschool-aged children (n = 250) to predict parent-reported child anxiety, a widely used and accessible indicator that guides parents’ decisions to seek treatment. Study 1 assessed the validity of a novel computer vision approach for assessing children’s behavioral risk for anxiety. A computer vision and machine learning prediction model (FEAR-CAM) was trained to assess children’s fear behaviors, an index of BI, during a modified stranger approach task. Validation analyses indicate strong model performance using both training-testing splits and cross-validation, with accuracy comparable to that of gold-standard behavioral coding methods. Study 2 examined whether and how physiological regulation informs reported anxiety using indicators of children’s moment-by-moment parasympathetic regulation (vagal flexibility) across the stranger approach task. Findings indicate variability in non-linear regulatory patterns, and associations with reported anxiety appear sensitive to analytic specification. Alternative approaches to modeling these links are explored. Study 3 examined 63 risk factors indexing multiple domains of childhood anxiety risk, including biobehavioral measures. Predictors were ranked using SHAP values derived from a trained XGBoost model to examine their contribution to model predictions, with results suggesting that child-level factors, particularly emotion symptoms, BI, and parasympathetic regulation, were the most influential contributors to parent-reported anxiety. Together, these findings highlight the potential of scalable behavioral and physiological assessment approaches, integrated with parent-reported indicators, to improve early identification of children at risk for anxiety."]},{"key":"dc:title","label":"Title","values":["Elucidating biobehavioral risk factors for anxiety in early childhood"]}]}],"canonical_facts":{"dc:contributor.advisor":["Wagner, Nicholas"],"dc:creator":["Sem, Kathy"],"dc:date.accessioned":["2026-10-15T17:37:23Z"],"dc:date.available":["2026-10-15T17:37:23Z"],"dc:date.issued":["2026"],"dc:date.updated":["2026-09-09T16:24:37Z"],"dc:description":["2026"],"dc:description.abstract":["Anxiety disorders are among the most common psychiatric conditions in childhood. Although behavioral inhibition (BI) is a robust predictor of anxiety risk, gold-standard observational assessments are time-intensive and difficult to scale, limiting their utility for early identification. Moreover, other important risk factors span many domains, including child characteristics, parent factors, dyadic relationships, and broader contexts (e.g., family conflict, household income), yet these are often studied in isolation. As a result, it remains unclear which risk markers contribute most strongly to children’s anxiety when considered together. To address this gap, the current dissertation integrates observational and physiological risk markers alongside commonly used parent-report measures from a large and diverse sample of preschool-aged children (n = 250) to predict parent-reported child anxiety, a widely used and accessible indicator that guides parents’ decisions to seek treatment. Study 1 assessed the validity of a novel computer vision approach for assessing children’s behavioral risk for anxiety. A computer vision and machine learning prediction model (FEAR-CAM) was trained to assess children’s fear behaviors, an index of BI, during a modified stranger approach task. Validation analyses indicate strong model performance using both training-testing splits and cross-validation, with accuracy comparable to that of gold-standard behavioral coding methods. Study 2 examined whether and how physiological regulation informs reported anxiety using indicators of children’s moment-by-moment parasympathetic regulation (vagal flexibility) across the stranger approach task. Findings indicate variability in non-linear regulatory patterns, and associations with reported anxiety appear sensitive to analytic specification. Alternative approaches to modeling these links are explored. Study 3 examined 63 risk factors indexing multiple domains of childhood anxiety risk, including biobehavioral measures. Predictors were ranked using SHAP values derived from a trained XGBoost model to examine their contribution to model predictions, with results suggesting that child-level factors, particularly emotion symptoms, BI, and parasympathetic regulation, were the most influential contributors to parent-reported anxiety. Together, these findings highlight the potential of scalable behavioral and physiological assessment approaches, integrated with parent-reported indicators, to improve early identification of children at risk for anxiety."],"dc:identifier.orcid":["0000-0002-5248-2561"],"dc:identifier.uri":["https://hdl.handle.net/2144/53698"],"dc:language.iso":["en_US"],"dc:rights":["Attribution-NonCommercial-NoDerivatives 4.0 International"],"dc:rights.uri":["http://creativecommons.org/licenses/by-nc-nd/4.0/"],"dc:subject":["Mental health","Physiological psychology","Behavioral psychology","Anxiety","Biobehavioral risk","Child anxiety risk","Computer vision","Machine learning","Physiological risk"],"dc:title":["Elucidating biobehavioral risk factors for anxiety in early childhood"],"dc:type":["Thesis/Dissertation"],"thesis:degree_discipline":["Psychological & Brain Sciences"],"thesis:degree_level":["doctoral"],"thesis:degree_name":["Doctor of Philosophy"],"thesis:institution_name":["Boston University"]},"updated_at":"2026-11-18T19:43:36Z"}