Eyes, Window to the Mind: Tracing the Structure of Cognitive Processing Stages in Associative Recognition using Pupil Dilatation Deconvolution

Nedelcu, Mara (2023) Eyes, Window to the Mind: Tracing the Structure of Cognitive Processing Stages in Associative Recognition using Pupil Dilatation Deconvolution. Bachelor's Thesis, Artificial Intelligence.

Preview	Text Cognitive_Processing_Stages_in_Associative_Recognition_using_Pupil_Dilation_Deconvolution.pdf Download (2MB) | Preview
	Text toestemming.pdf Restricted to Registered users only Download (130kB)

Abstract

Associative Recognition(AR) is a cognitive task providing insight into the stages required for information processing. While studied from multiple perspectives, there is still no clear definition for the exact structure of its processing stages. More so, the manner in which the intensity level of mental effort impacts different stages has not yet been dealt with. It is known that the human pupil provides information regarding cognitive processes, its size increasing with mental effort. Here we applied a new approach employing Hidden semi-Markov Model and Generalized Additive Mixed Model(HsMM-GAMM) to perform pupillary deconvolution on the data from an AR experiment. Based on this, we proposed an additional model of AR. Results show a significant effect of fan on both response time(RT) and error rate(ER), with an effect of probe type only significant on RT. Employing HsMM-GAMMs in the pupil response analysis for a new AR model provides insight into the variability and duration of the processing stages of the cognitive task through the lens of mental effort. The most probable resulting model comprises six stages and resembles a previously proposed MEG model. While recovering a sensible stage structure, the HsMM-GAMM model is not without fault. The greatest downsides of the architecture were its lack of specificity, influencing its assignment of intensity level of mental effort to specific stages, its modelling of some individual stage responses as having negative amplitudes.

Item Type:	Thesis (Bachelor's Thesis)
Supervisor name:	Borst, J.P. and Krause, J.
Degree programme:	Artificial Intelligence
Thesis type:	Bachelor's Thesis
Language:	English
Date Deposited:	31 Jul 2023 07:33
Last Modified:	31 Jul 2023 07:33
URI:	https://fse.studenttheses.ub.rug.nl/id/eprint/30968

Actions (login required)

View Item