The masculine overcompensation thesis asserts that men react to masculinity threats with extreme demonstrations of masculinity, a proposition tested here across four studies. In study 1, men and women were randomly given feedback suggesting they were either masculine or feminine. Women showed no effects when told they were masculine; however, men given feedback suggesting they were feminine expressed more support for war, homophobic attitudes, and interest in purchasing an SUV. Study 2 found that threatened men expressed greater support for, and desire to advance in, dominance hierarchies. Study 3 showed in a large-scale survey on a diverse sample that men who reported that social changes threatened the status of men also reported more homophopic and prodominance attitudes, support for war, and belief in male superiority. Finally, study 4 found that higher testosterone men showed stronger reactions to masculinity threats than those lower in testosterone. Together, these results support the masculine overcompensation thesis, show how it can shape political and cultural attitudes, and identify a hormonal factor influencing the effect.

How do minds produce explicit attitudes over several hundred milliseconds? Speeded evaluative measures have revealed implicit biases beyond cognitive control and subjective awareness, yet mental processing may culminate in an explicit attitude that feels personally endorsed and corroborates voluntary intentions. We argue that self-reported explicit attitudes derive from a continuous, temporally dynamic process, whereby multiple simultaneously conflicting sources of information self-organize into a meaningful mental representation. While our participants reported their explicit (like vs. dislike) attitudes toward White versus Black people by moving a cursor to a "like" or "dislike" response box, we recorded streaming x- and y-coordinates from their hand-movement trajectories. We found that participants' hand-movement paths exhibited greater curvature toward the "dislike" response when they reported positive explicit attitudes toward Black people than when they reported positive explicit attitudes toward White people. Moreover, these trajectories were characterized by movement disorder and competitive velocity profiles that were predicted under the assumption that the deliberate attitudes emerged from continuous interactions between multiple simultaneously conflicting constraints.

Sophisticated malware authors can sneak hidden malicious contents into portable executable files, and these contents can be hard to detect, especially if encrypted or compressed. However, when an executable file switches between content regimes (e.g., native, encrypted, compressed, text, and padding), there are corresponding shifts in the file's representation as an entropy signal. In this paper, we develop a method for automatically quantifying the extent to which patterned variations in a file's entropy signal make it "suspicious." In Experiment 1, we use wavelet transforms to define a Suspiciously Structured Entropic Change Score (SSECS), a scalar feature that quantifies the suspiciousness of a file based on its distribution of entropic energy across multiple levels of spatial resolution. Based on this single feature, it was possible to raise predictive accuracy on a malware detection task from 50.0% to 68.7%, even though the single feature was applied to a heterogeneous corpus of malware discovered "in the wild." In Experiment 2, we describe how wavelet-based decompositions of software entropy can be applied to a parasitic malware detection task involving large numbers of samples and features. By extracting only string and entropy features (with wavelet decompositions) from software samples, we are able to obtain almost 99% detection of parasitic malware with fewer than 1% false positives on good files. Moreover, the addition of wavelet-based features uniformly improved detection performance across plausible false positive rates, both in a strings-only model (e.g., from 80.90% to 82.97%) and a strings-plus-entropy model (e.g. from 92.10% to 94.74%, and from 98.63% to 98.90%). Overall, wavelet decomposition of software entropy can be useful for machine learning models for detecting malware based on extracting millions of features from executable files. 1

In this article we consider the phenomenon of evaluative readiness, whereby the activation in memory of a goal leads to an unintentional increase in positivity toward stimuli that can facilitate the goal. We review four lines of work that together address the question of when goals lead to this kind of automatic shift in people's attitudes. We then consider how contemporary models of cognition might explain this effect. We review whether dual systems models and single interacting system models can explain the phenomenon of evaluative readiness. Based on recent work in cognitive psychology and computational neuroscience, we then argue for the potential explanatory value of turning to a multiple interacting systems framework for explaining the phenomenon of evaluative readiness.

Malicious software ('malware')