Perceptual Decision Making in Rodents

<p dir="ltr">Perceptual decision making involves gathering evidence and integrating information about sensory stimuli in the world. This has been studied in many ways across species and sensory modalities, with many studies utilizing the Drift Diffusion Model (DDM) (Ratcliff et al., 2016) to better understand the cognitive processes involved in decision making. Drift rate is a key parameter of the DDM. It represents how quickly animals integrate sensory information, which is impacted by factors such as task difficulty and attention (Myers et al., 2022). Another important parameter is called threshold, which represents the amount of evidence or information needed to make a choice, which might be influenced by the level of caution (Myers et al., 2022). Related to threshold is the starting point bias parameter, which affects the frequency of decisions to a particular response, for instance if one response is more highly rewarded than another (Myers et al., 2022). Therefore, parameters of DDMs seem to be impacted by both stimulus properties and rewards. One goal of the studies presented in this dissertation is to understand how stimulus properties versus reward values impact perceptual decision making behavior and the parameters of the DDM in rats. To do so, we used visual cues of different luminance levels, as visual decision making studies are historically less common but have become increasingly popular in rodent models (Zoccolan et al., 2009; Busse et al., 2011). In our experiments, brighter cues were first associated with larger reward values, and dimmer cues with smaller reward values (Chapter 2). We then swapped either the brighter or dimmer cue with an intermediate cue, while keeping reward values fixed (Chapter 3). To study the specific effects of cue luminance, we presented animals with cues of different luminance levels which were all rewarded equally (Chapter 4). Our behavioral evidence suggests that choice accuracy depends on cue luminance, which is accompanied by effects on drift rate. Response time depended on both cue luminance and reward value, and was coupled with effects on threshold and starting point bias. A second goal of this dissertation was to investigate the role of the prelimbic cortex (PLC), part of the rodent medial prefrontal cortex (mPFC), in perceptual decision making. To do so we used reversible inactivations (Chapter 2) and electrophysiological recordings (Chapter 5). Inactivating PLC resulted in faster responses coupled with a reduction in threshold, indicating the PLC’s role in inhibitory processes and response caution. Additionally, we observed coherence in low frequency neural signals when animals responded quickly and correctly. Taken together, these findings suggest the PLC is involved in guiding the actions to make decisions, which is reflected behaviorally in response speed.</p>