Losing control, literally: Relations between anger control, trait anger, and motor control

Abstract

Self-control perspectives of multiple traits have been proposed, perhaps most particularly so in the anger realm. Four studies sought to examine potential relations between anger control, trait anger, and motor control. Across the four studies, individuals (total N=366) were asked to hold a joystick cursor on a spatial target as accurately and steadily as possible and two indices of motor control were quantified. Studies 1 and 2 found that higher levels of (trait) anger control were predictive of better motor control. Studies 3 and 4 then showed that higher levels of trait anger were predictive of worse motor control. All studies also examined possible state-related influences on motor control (e.g., as a function of aversive noise), but no such effects were found. Thus, the trait-related findings were basic in nature and informative for this reason. Results are discussed in terms of their implications for understanding personality variations in anger control and anger and the value of motoric probes of self-control.

Keywords:

• Anger control
• Trait anger
• Self-control
• Motor control
• Implicit

Acknowledgments

The authors acknowledge support from NSF (BCS 0843982).

Notes

¹Due to a clerical error, one anger control item was not presented. The remaining anger control items constituted a reliable scale.

²Reaction times to engage with the task—i.e., to move the cursor to the target—did not systematically vary by anger control or trait anger in any of the studies, all ps>.05. We can therefore infer that the findings implicate basic differences in the ability to control one's behaviour in moment-to-moment terms rather than differential engagement with the task.

³There was no main effect for sex in any of the analyses and this was true for both metrics for all studies, all ps > .05. For each of the two dependent measures, for each of the studies, we also performed a multiple regression in which the trait variable (e.g., anger control) and a dummy-coded sex variable were entered as simultaneous predictors of motor control. The significant simple regression results reported in the text remained significant when controlling for sex, all ps<.05.

⁴We also examined potential interactions between the trait variables and the priming manipulations using general linear modelling (GLM) procedures. The trait variable was z-scored (Robinson, 2007b) and crossed with the priming manipulation. In no case were there trait by manipulation interactions and this was true for both measures for all studies, all ps>.05. These results further suggest that motor control is an individual difference variable that is relatively insensitive to state-related factors, at least as manipulated.