Rapid encoding of task regularities in the human hippocampus guides sensorimotor timing

Abstract

The brain encodes the statistical regularities of the environment in a task-specific yet flexible and generalizable format. Here, we seek to understand this process by bridging two parallel lines of research, one centered on sensorimotor timing, and the other on cognitive mapping in the hippocampal system. By combining functional magnetic resonance imaging (fMRI) with a fast-paced time-to-contact (TTC) estimation task, we found that the hippocampus signaled behavioral feedback received in each trial as well as performance improvements across trials along with reward-processing regions. Critically, it signaled performance improvements independent from the tested intervals, and its activity accounted for the trial-wise regression-to-the-mean biases in TTC estimation. This is in line with the idea that the hippocampus supports the rapid encoding of temporal context even on short time scales in a behavior-dependent manner. Our results emphasize the central role of the hippocampus in statistical learning and position it at the core of a brain-wide network updating sensorimotor representations in real time for flexible behavior.

Context-specific and context-invariant computations of interval timing

“…In the discrimination task, subjects had to measure and compare two different sample intervals, ts1 and ts2, demarcated by three visual flashes. The values of ts1 were selected from a discrete uniform distribution, identical to the value of ts in the reproduction task, and the values of ts2 were chosen from the distribution of ts1 ± [6,12,24,48] % of ts1, randomly on each trial. Subjects compared ts2 with ts1 and chose whether ts2 was longer or shorter than ts1 via a manual key press or a saccadic eye movement.…”

“…Duration of ts2 was ts1 duration ± [6,12,24,48]% of ts1 duration, selected pseudorandomly on each trial. Feedback was shown at the end of each trial for 0.8 sec (a green circle with a 1.25° diameter for correct trials and a red circle of the same size for incorrect trials).…”

“…However, electrophysiological and computational studies showed that time can be encoded through changes in neural population activity over time, or population state dynamics [16][17][18][19][20] . Although dedicated and intrinsic models of timing, are not mutually exclusive [21][22][23][24] , converging data from psychophysical, neuroimaging and electrophysiological studies supported partially overlapping distributed timing mechanisms for interval timing (for a review see 1,25,26 ). Indeed, A longstanding question is how these distributed timing circuits interact to shape temporal behavior.…”

Mental search of concepts is supported by egocentric vector representations and restructured grid maps

“…To answer our question we used the recently developed DeepMREye toolbox 35 , a convolutional neural network that decodes gaze position from the magnetic resonance signal of the eyeballs. We used a set of pre-trained weights, freely available from the toolbox website and resulting from previous experiments where concurrent BOLD and eye-tracking signals were recorded [36][37][38] , to obtain the estimates of gaze position for each fMRI volume, in each run and subject. First, we wanted to make sure that the tool could be reliably applied to our dataset.…”

Encoding of task regularities links grid-like signals to human timing behavior

“…An emerging view is that the hippocampal formation plays an important role in encoding regularities that afford generalization across tasks or contexts (e.g., different environments; Whittington et al (2020); Bousquet et al (1998); Fuhs, Touretzky (2007); Penny et al (2013); Friston, Buzsáki (2016); Pezzulo et al (2017)), thus greatly accelerating learning and reducing behavioral errors in novel or noisy situations (Lisman, Redish, 2009; Stachenfeld et al, 2017). In particular, the hippocampus has been suggested to support the encoding of task regularities in real time as a task is performed, as its activity reflects feedback and behavioral improvements even in fast-paced timing tasks (Polti et al, 2022).…”