Illusions are a powerful tool for studying the single neuron correlates of perception. Here, we introduce the neon color spreading (NCS) illusion in mice and report the neuronal correlates of illusory brightness, which has heretofore only been studied using human fMRI. We designed a novel NCS paradigm to evoke the percept of an illusory drifting grating and analyzed the activity of 520 single units in the mouse primary visual cortex (V1). A substantial proportion of V1 single units (60.5%) responded to illusory gratings with direction tuning matched to their preferred direction, which was determined using physically presented luminance-defined gratings (LDG). Moreover, by presenting LDG gratings with a 180 ° phase shift relative to NCS gratings, we show that spatial phase tuning shifted 180 ° for most single units. This finding conclusively demonstrates that V1 single units respond to illusory brightness. Using this novel mouse paradigm, we show that responses to illusory gratings have a lower magnitude and are delayed relative to physical gratings. We determined where V1 single units fell in the V1 cellular hierarchy (based on their susceptibility to surround suppression, their putative classification as interneuron or pyramidal neuron, and designation as a simple or complex cell) and found that higher-level V1 single units are more responsive to NCS stimuli. These findings resolve the debate of whether V1 is involved in illusory brightness processing and reveal a V1 hierarchical organization in which higher-level neurons are pivotal to the processing of illusory qualities, such as brightness.

Re-exposure to the context that information was learned in facilitates its memory retrieval. However, the influence of context changes on the ability to learn new information is less well understood, which the present work investigated in two experiments with healthy participants (n = 40 per experiment; 20 female). In experiment 1, participants learned a list of word-pairs (A-B) in the morning, after which their memory for the word-pairs was immediately tested. In the evening, they learned and were tested on a second non-overlapping list (C-D), either in the same context or in a different context than the first list (between-subjects). We found that new learning is enhanced in the same context, and that new learning in the other context was decreased compared to baseline. In experiment 2, participants were exposed to both contexts in the morning, but only learned word-pairs in one of them. In the second learning session in the evening, this familiarization with the other context abolished differences between the same and other context group. These data point to context novelty interfering with new learning rather than context familiarity enhancing it. Importantly, the reduction of new learning in the other context in the first experiment, where the context was unfamiliar in both learning sessions, suggests mechanisms beyond attention processes that are bound by the novelty of the other context. Rather, the old context impairs the processing of the new context, possibly by biasing pattern completion and pattern separation trade-offs within the hippocampus.

Context plays a key role in learning and memory processes. Re-exposure to the context that information was learned in facilitates memory retrieval of this information. However, it is currently unclear whether context changes also influence the ability to learn new information, which the present work investigated in two experiments with healthy participants (n = 40 per experiment; 20 female). In experiment 1, participants learned a list of word-pairs (A-B) in the morning and a second non-overlapping list (C-D) in the evening, either in the same context or in a different context than the first list (between-subjects). We confirmed that new learning is enhanced if it takes place in the same context, putatively driven by context-dependent retrieval of meta-learning processes. In addition, new learning in the other context was significantly decreased compared to baseline. In experiment 2, participants were exposed to both contexts in the morning, but only learned word-pairs in one of them. Familiarity with the other context abolished differences between the same and other context group. These data point to the novelty of the context interfering with new learning rather than the familiarity of the context enhancing it. Importantly, the reduction of new learning in the other context in the first experiment, where the context was unfamiliar in both learning sessions, suggests mechanisms beyond attention processes that are bound by the novelty of the other context. Rather, the old context impairs the processing of the new context, possibly by biasing pattern completion and pattern separation trade-offs within the hippocampus.