| Abstract: | The process by which sensory evidence contributes to perceptual choices requires an understanding of its transformation into decision variables. Here, we address this issue by evaluating the neural representation of acoustic information in the auditory cortex-recipient parietal cortex, while gerbils either performed a two-alternative forced-choice auditory discrimination task or while they passively listened to identical acoustic stimuli. During task engagement, stimulus identity decoding performance from simultaneously recorded parietal neurons significantly correlated with psychometric sensitivity. In contrast, decoding performance during passive listening was significantly reduced. Principal component and geometric analyses revealed the emergence of low-dimensional encoding of linearly separable manifolds with respect to stimulus identity and decision, but only during task engagement. These findings confirm that the parietal cortex mediates a transition of acoustic representations into decision-related variables. Finally, using a clustering analysis, we identified three functionally distinct subpopulations of neurons that each encoded task-relevant information during separate temporal segments of a trial. Taken together, our findings demonstrate how parietal cortex neurons integrate and transform encoded auditory information to guide sound-driven perceptual decisions.Integrating sensory information over time is one of the fundamental attributes that is required for accurate perceptual decisions (1, 2). This process is supported by the transformation of stimulus representations into decision variables. In the case of auditory stimuli, prior to the formation of decision variables, the central representations of acoustic cues are gradually reconfigured along the auditory neuraxis. Thus, auditory neurons become more selective to contextually relevant acoustic features as one ascends the central pathway into the auditory cortex (3). Ultimately, individual acoustic components merge into auditory objects to guide perception (4). Similarly, primary visual cortex neurons are selective to the stimulus orientation (5, 6), whereas higher cortices are selective for more complex characteristics (7–9). A hierarchical progression of sensory information processing is also seen across the somatosensory ascending pathway where receptive fields grow more complex (10). These hierarchically transformed neural signals are ultimately decoded downstream of sensory cortices for stimulus-dependent decisions (4, 11–14).Studies in both nonhuman primates and rodents suggest that the parietal cortex integrates sensory inputs and transforms them into decision signals (15–19). The parietal cortex receives direct projections from primary or secondary sensory cortices (20, 21), has been causally implicated in the performance of perceptual decision-making tasks (22–25), and its activity typically reflects action selection (26, 27). Furthermore, parietal neurons gradually increase their spiking activity over time epochs that scale with the accumulation of sensory evidence (11, 28–31). Thus, while parietal cortex activity reflects decision variables, the manner in which relevant sensory stimuli are represented prior to this transformation remains uncertain.To dissociate encoding of stimuli from encoding of decision, we recorded neural activity from the parietal cortex while gerbils performed an auditory discrimination task (25), and again during passive listening sessions, using the same acoustic stimuli in the absence of behavioral decision. While some visual studies have explored visual selectivity of parietal cortex neurons under passive fixation conditions (32, 33), a direct comparison between the decoding of visual stimuli versus decision would require that eye fixation be controlled during stimulus presentation. In contrast, auditory tasks can be performed without the need to maintain head position during a trial, permitting us to directly compare the sound-driven responses of parietal cortex neurons during task engagement versus their responses to identical stimuli during passive listening. Thus, we predicted that if parietal cortex activity during task performance did not reflect the transition into decision-related variables, then all analyses of neural processing would be similar to those displayed during the passive listening condition. We found that during task performance, decoded parietal cortex population activity based on stimulus identity correlated with behavioral discrimination across similar timescales. Furthermore, principal component analysis (PCA) performed on parietal cortex responses revealed neural trajectories (i.e., the change in parietal cortex population activity over time) that demonstrated the temporal progression of low-dimensional encoding of acoustic information that transitioned to encoding of behavioral choices. During passive listening sessions, decoding performance from parietal cortex population activity based on stimulus identity was poorer than decoding during task performance, but scaled with stimulus duration. In addition, the PCA revealed neural trajectories that differentiated between each stimulus condition, but did not reflect a decision variable. Thus, the parietal cortex could accumulate auditory evidence for the purpose of forming a decision variable during task performance. Finally, our clustering analysis based on neuronal response properties suggest subpopulations of parietal neurons that may reflect separate temporal segments of individual trials during decision-making. We propose that the parietal cortex integrates and transforms bottom-up sensory information into decision variables during task performance. |
