Prediction strength modulates responses in human area CA1 to sequence violations

Abstract

Emerging human, animal, and computational evidence suggest that, within the hippocampus, stored memories are compared with current sensory input to compute novelty, i.e., detecting when inputs deviate from expectations. Hippocampal subfield CA1 is thought to detect mismatches between past and present, and detected novelty is thought to modulate encoding processes, providing a mechanism for gating the entry of information into memory. Using high-resolution functional MRI, we examined human hippocampal subfield and medial temporal lobe cortical activation during prediction violations within a sequence of events unfolding over time. Subjects encountered sequences of four visual stimuli that were then reencountered in the same temporal order (Repeat) or a rearranged order (Violation). Prediction strength was manipulated by varying whether the sequence was initially presented once (Weak) or thrice (Strong) prior to the critical Repeat or Violation sequence. Analyses of blood oxygen level-dependent signals revealed that task-responsive voxels in anatomically defined CA1, CA23/dentate gyrus, and perirhinal cortex were more active when expectations were violated than when confirmed. Additionally, stronger prediction violations elicited greater activity than weaker violations in CA1, and CA1 contained the greatest proportion of voxels displaying this prediction violation pattern relative to other medial temporal lobe regions. Finally, a memory test with a separate group of subjects showed that subsequent recognition memory was superior for items that had appeared in prediction violation trials than in prediction confirmation trials. These findings indicate that CA1 responds to temporal order prediction violations, and that this response is modulated by prediction strength.

Transcending time in the brain: How event memories are constructed from experience

“…A separate fMRI study showed that the hippocampal CA1 subregion is sensitive to temporal context information, such that hippocampal representations of the same target item differed depending on whether it was preceded by its two original images or two slightly similar images (Wang & Diana, ). This finding accords with other work showing that different hippocampal sub‐regions respond to temporal sequence (Chen, Cook, & Wagner, ; Kim et al, 2017) and temporal order violations (Azab, Stark, & Stark, ), as well as expected stimulus violations more broadly (Duncan, Ketz, Inati, & Davachi, ). Furthermore, the hippocampus has also been shown to be sensitive not only to item violations but also temporal duration violations; however, in this instance, hippocampal activity was greater when the expected item appeared (Barnett et al, ).…”

Hippocampal Mismatch Signals Are Modulated by the Strength of Neural Predictions and Their Similarity to Outcomes

“…Finally, hippocampal responses to unexpected outcomes were associated with subsequent behavioral expressions of memory updating, with greater hippocampal out- come responses predicting relatively faster retrieval of new associations and relatively slower retrieval of older associations. Compared with prior reports of hippocampal mismatch signals Maguire, 2006a, 2006b;Duncan et al, 2009Duncan et al, , 2012Chen et al, 2015), a critical-and novel-advantage of our experimental approach is that we measured directly, on a trialby-trial basis, neural prediction strength. Although the hippocampus is known to act as a novelty detector (Stern et al, 1996;Strange et al, 1999;Ranganath and Rainer, 2003), the mismatch signal is not thought to simply reflect associative novelty, but to reflect a comparison between an actively generated prediction and a new, but unexpected associative outcome (Kumaran and Maguire, 2007).…”

“…Therefore, the hippocampus may play a critical role both in generating predictions and comparing predictions to outcomes (Hasselmo and Wyble, 1997;Lisman and Grace, 2005;Kumaran and Maguire, 2007;Chen et al, 2015). That said, when we decoded prediction strength directly from the hippocampus, it was not related to univariate hippocampal outcome responses.…”

“…Many theoretical perspectives (Gluck and Myers, 1993;Eichenbaum, 2004;Lisman and Grace, 2005;Rolls, 2013;Davachi and DuBrow, 2015) and empirical findings (Kumaran and Maguire, 2006a;Duncan et al, 2009;Chen et al, 2011) suggest that the hippocampus plays a critical role in comparing memory-based predictions with perception-based outcomes. When predictions do not match outcomes, the hippocampus is thought to generate mismatch signals that reflect these expectancy violations (Kumaran and Maguire, 2006b;Duncan et al, 2012;Chen et al, 2015).…”

“…Although mismatch signals should depend critically on the active generation of predictions (Kumaran and Maguire, 2007), existing evidence for hippocampal mismatch signals comes from paradigms in which prediction strength is inferred but not measured directly (Kumaran and Maguire, 2006b;Chen et al, 2015). Moreover, an important but unresolved question is whether hippocampal mismatch signals are sensitive to the degree of similarity between predictions and outcomes.…”