Social cognition deficits are observed both in patients with schizophrenia (SCZ) and in patients with mesial temporal lobe epilepsy (MTLE). This may be due to dysfunction of the amygdala network, which is a common feature of both diseases. In this study, SCZ (n = 48) or MTLE (n = 31) and healthy controls (HC, n = 47) completed assessments of mentalising (Reading Mind in the Eyes Test, RMET) and basic cognitive processing, e.g., working memory, executive functions and psychomotor speed (Trail-Making Test B and Digit Symbol). SCZ were also assessed with the Positive And Negative Syndrome Scale (PANSS). We found that the RMET scores of the two clinical groups were similar (p > 0.05) and lower than in the HCs (SCZ: p < 0.05; MTLE: p < 0.001). In the next step, SCZ were split into two groups with respect to the level of symptoms. Analysis of the RMET scores revealed no differences between the HC (M = 25.7 ± 4.1) and POS-LO (M = 25.3 ± 4.8); both groups outperformed the POS-HI group (M = 21.3 ± 5.2) and the MTLE group (M = 20.8 ± 4.6). No differences were found for the median-split with regard to negative symptoms. In SCZ, the mind-reading deficit appears to be associated with the level of positive symptoms. Both POS-HI and MTLE patients present significant mentalising deficits compared to healthy controls.

No abstract

Trying to guess what the correct answer to a question might be can facilitate future learning of this answer when presented in the form of corrective feedback. One issue that determines the effectiveness of guessing as a learning strategy is the timing of the presentation of feedback: it can be presented either immediately after the guess, or after a delay. Whereas the timing of feedback is of little importance for complex materials such as trivia questions, previous research suggests that for simpler materials such as related word pairs guessing seems to benefit learning only when feedback is immediate. In order to test whether this always has to be the case, we conducted two experiments in which we increased the richness of study materials by superimposing the to-be-learned word pairs over unrelated context pictures. We then manipulated the match between contexts at study and at test (Experiment 1) and at the time of feedback delivery (Experiment 2). Contrary to previous studies showing no benefits of guessing with delayed feedback, our results show that learning related word pairs can benefit from guessing even when feedback is delayed. These benefits of guessing occur if participants are reminded via reinstated contexts of the guessing stage at the time of feedback delivery. Our results help constrain theories of guessing benefits and extend theories of reminding.