Investigating the pupil response as a physiological correlate of encoding and retrieval

Thesis

The question of how we encode and retrieve previously learned information has been explored for millennia. However, many unanswered questions remain. For instance, in daily life these two processes are constantly interleaved, and yet research has shown that they require opposing neural mechanisms. Therefore, what facilitates this ability to rapidly switch back and forth between encoding and retrieval? An influential model by Hasselmo (1995) suggests that acetylcholine is a key candidate for modulating this process. This theory has been extensively studied in animal and computational models but work using human participants has been scarce. The first goal of this thesis project was to address this gap.

To do so, we employed pupillometry in a continuous associative memory paradigm with alternating encoding and retrieval tasks. Changes in the pupil size are indirect markers of neuromodulatory activity especially that from the locus coeruleus - norepinephrine (LC- NE) and acetylcholine. Additionally, pupillometry has increasingly become touted as a reliable tool for measuring cognitive processing. However, there is debate as to what exactly pupil dilation means in the context of recognition memory. Does it reflect memory strength or the cognitive effort exerted to retrieve relevant traces? Furthermore, there is also debate as to whether pupillary changes do in fact index cognitive processes or if they simply capture mere task engagement. The second and third goal of the current study was to address these debates, respectively.

So, what did we find? First, even when task requirements were matched, encoding and retrieval exhibited distinct pupil patterns. This provided evidence for the differential impact of acetylcholine and norepinephrine on these processes which is consistent with Hasselmo’s predictions. Second, the pupil time course for retrieval was comprised of two phases: an early and later component whose characteristics could be best explained by the memory strength and retrieval effort account, respectively. Thus, allowing for a reconciliation of the mixed findings in the literature. Lastly, the pupil pattern replicated three of the most consistent findings in memory studies, namely, the old/new effect, the encoding specificity principle, and the observation that memories decay as time passes. The fact that we could replicate these well-established memory effects suggests that the pupil response is indeed sensitive to cognitive processing.

Authors: Gloria, C

• DOI: 10.5287/ora-eyro16rmn
• Type of award: MSc by Research
• Level of award: Masters
• Awarding institution: University of Oxford
• Language: English
• Keywords: memory retrieval; neuromodulation; eye tracking; pupillometry; memory encoding; episodic memory
• Subjects: Episodic memory; Pupillometry