Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Discovery Grants Program - Individual

University of Alberta

$51,000.00

Basic mechanisms of directionality and association in relational memory

FY 2023-24

Summary
The power of human memory is in reconstructing not only what items (such as words) were studied, but their relationships. The basic building blocks of relational memory are associations between pairs of items. We aim to advance theory and experimental approaches to understand the cognitive basis of relational memory by tackling recent challenges to mathematical models of association. 1) Relational influence. One item can influence which features of a paired item are attended; RIVER evokes different features of BANK than does VAULT. By letting RIVER and VAULT (for example) each cause a different subset of features of BANK to be encoded, our mathematical model could explain preserved associative memory in amnesia. Series 1a tests model-predictions that relational influence can improve associative recognition (were RIVER and BANK studied together?) while being detrimental to cued recall (given BANK, recall the paired word) and item recognition (was BANK on the list?). Series 1b will test whether we can manipulate relational influence to both enhance and reduce the classic finding, recognition-failure of recallable words. Series 1c will use machine-learning classifiers to predict memory outcome based on study-related EEG activity. We will test the hypothesis that features that classify associative recognition will inversely predict item-recognition and cued recall when relational influence is high. By evaluating the amount and nature of relational influence in intact-participant memory, Series 1 will tell us how to develop models to accurately incorporate such meaning-driven effects. 2) Order within associations. My lab found that participants remember associations with imperfect memory for constituent-order (order recognition: was it WALL BABY or BABY WALL?). Our models could do this by adding item-order associations, coding order as item features or incorporating order in the association itself. Series 2a will test the hypothesis that models that provide order by coding the position of items separately may be preferred because they are cognitively less demanding, but will fail in certain situations (lists like AB, BC, where B is both a left- and right-hand item). In the latter case, participants may be storing order within the association itself. Series 2b will build models of longer ordered lists from the association models. We will test whether any of these so-called chaining models will be able to explain some important benchmark findings by using associations that are symmetric in strength (equal probability of going from BABY to WALL as from WALL to BABY), while still providing some way to judge order. Series 2 will develop a more complete theory of relational memory from associations to serial lists. These studies will produce new theory and data that will advance our knowledge of the cognitive and neural processes that support relational memory, with the potential to guide memory assessment and training protocols.
Program
Discovery Grants Program - Individual
Location
Alberta, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data

As published

Application id
RGPIN-2023-04196
Name nom
Caplan, Jeremy
Department département
Psychology
Organization id
9
Institution établissement
University of Alberta
Province en
Alberta
Province fr
Alberta
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2023
Competition year année de concours
2023
Award amount
51000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1502
Committee name en
Biological Systems and Functions
Committee name fr
Systèmes et fonctions biologiques
Area of application code
1210
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Psychology
Area of application fr
Psychologie
Research subject code
9999
Research subject group en
Not available
Research subject group fr
Not available
Research subject en
Non disponible
Research subject fr
Non disponible
Application title
Basic mechanisms of directionality and association in relational memory
Keyword
Classifiers; Cognitive Neuroscience; Cognitive Psychology; Episodic memory; Human Behaviour; Mathematical Models; Memory; Memory for associations; Memory for order; Subsequent memory effect
Application summary
The power of human memory is in reconstructing not only what items (such as words) were studied, but their relationships. The basic building blocks of relational memory are associations between pairs of items. We aim to advance theory and experimental approaches to understand the cognitive basis of relational memory by tackling recent challenges to mathematical models of association. 1) Relational influence. One item can influence which features of a paired item are attended; RIVER evokes different features of BANK than does VAULT. By letting RIVER and VAULT (for example) each cause a different subset of features of BANK to be encoded, our mathematical model could explain preserved associative memory in amnesia. Series 1a tests model-predictions that relational influence can improve associative recognition (were RIVER and BANK studied together?) while being detrimental to cued recall (given BANK, recall the paired word) and item recognition (was BANK on the list?). Series 1b will test whether we can manipulate relational influence to both enhance and reduce the classic finding, recognition-failure of recallable words. Series 1c will use machine-learning classifiers to predict memory outcome based on study-related EEG activity. We will test the hypothesis that features that classify associative recognition will inversely predict item-recognition and cued recall when relational influence is high. By evaluating the amount and nature of relational influence in intact-participant memory, Series 1 will tell us how to develop models to accurately incorporate such meaning-driven effects. 2) Order within associations. My lab found that participants remember associations with imperfect memory for constituent-order (order recognition: was it WALL BABY or BABY WALL?). Our models could do this by adding item-order associations, coding order as item features or incorporating order in the association itself. Series 2a will test the hypothesis that models that provide order by coding the position of items separately may be preferred because they are cognitively less demanding, but will fail in certain situations (lists like AB, BC, where B is both a left- and right-hand item). In the latter case, participants may be storing order within the association itself. Series 2b will build models of longer ordered lists from the association models. We will test whether any of these so-called chaining models will be able to explain some important benchmark findings by using associations that are symmetric in strength (equal probability of going from BABY to WALL as from WALL to BABY), while still providing some way to judge order. Series 2 will develop a more complete theory of relational memory from associations to serial lists. These studies will produce new theory and data that will advance our knowledge of the cognitive and neural processes that support relational memory, with the potential to guide memory assessment and training protocols.
University of Alberta - $51,000 Grant (2023-24) | Canada Spends