Differential Effects of Endogenous and Exogenous Attention on Sensory Tuning.

Covert spatial attention (without concurrent eye movements) improves performance in many visual tasks (e.g., orientation discrimination and visual search). However, both covert attention systems-endogenous (voluntary) and exogenous (involuntary)-exhibit differential effects on performance in tasks mediated by spatial and temporal resolution suggesting an underlying mechanistic difference. We investigated whether these differences manifest in sensory tuning by assessing whether and how endogenous and exogenous attention differentially alter the representation of two basic visual dimensions-orientation and spatial frequency (SF). The same human observers detected a grating embedded in noise in two separate experiments (with endogenous or exogenous attention cues). Reverse correlation was used to infer the underlying neural representation from behavioral responses, and we linked our results to established neural computations via a normalization model of attention. Both endogenous and exogenous attention similarly improved performance at the attended location by enhancing the gain of all orientations without changing tuning width. In the SF dimension, endogenous attention enhanced the gain of SFs above and below the target SF, whereas exogenous attention only enhanced those above. Additionally, exogenous attention shifted peak sensitivity to SFs above the target SF, whereas endogenous attention did not. Both covert attention systems modulated sensory tuning via the same computation (gain changes). However, there were differences in the strength of the gain. Compared with endogenous attention, exogenous attention had a stronger orientation gain enhancement but a weaker overall SF gain enhancement. These differences in sensory tuning may underlie differential effects of endogenous and exogenous attention on performance.SIGNIFICANCE STATEMENT Covert spatial attention is a fundamental aspect of cognition and perception that allows us to selectively process and prioritize incoming visual information at a given location. There are two types: endogenous (voluntary) and exogenous (involuntary). Both typically improve visual perception, but there are instances where endogenous improves perception but exogenous hinders perception. Whether and how such differences extend to sensory representations is unknown. Here we show that both endogenous and exogenous attention mediate perception via the same neural computation-gain changes-but the strength of the orientation gain and the range of enhanced spatial frequencies depends on the type of attention being deployed. These findings reveal that both attention systems differentially reshape the tuning of features coded in striate cortex.

In smiles we trust? Smiling in the context of antisocial and borderline personality pathology.

Research suggests that people behave more cooperatively towards those who smile and less cooperatively towards those with personality pathology. Here, we integrated these two lines of research to model the combined effects of smiles and personality pathology on trust. In two experiments, participants read vignettes portraying a person with either borderline personality disorder, antisocial personality disorder, or no personality pathology. These portrayals were paired with a brief video clip that showed a person with either a neutral expression or a smile. Participants then played a Trust game with the "person" presented using each vignette and video clip combination. In Experiment 1, rates of trust were lower in response to the borderline and antisocial personality disorder vignettes compared with the control vignette. Interestingly, the effect of smiles was dependent upon personality. Although participants were more trusting of smiling confederates portrayed as having borderline personality disorder or no pathology, they were less trusting of confederates portrayed as having antisocial personality disorder if they smiled. In Experiment 2, run with a second set of personality vignettes, rates of trust were lower in response to both personality disorder vignettes and higher in response to smiles with no significant interaction. Together, these results suggest that information regarding both the current emotional state as well as the personality traits of a partner are important for creating trust.