Self-initiating stimulus in Short Term Memory: differential effects of temporal prediction and motor control

Classic studies of Short Term Memory (STM) required passively waiting for the stimulus. In contrast, sensory activation is typically the result of self-initiated movements in natural behavior. It has previously shown that self-initiation improves accuracy in an STM task (Loyola et al., 2022). However, it is still unknown whether this effect is related to the temporal predictability or to the activation of motor systems produced during self-initiation (motor control). We hypothesize that both mechanisms have independent effects on STM accuracy. Undergraduate students underwent a modified Stenberg task, which was designed to manipulate the degree of motor control and temporal predictability of the stimulus onset. We created five conditions: Active Predictable (AP), Active Unpredictable (AU), Passive Predictable (PP), Passive Unpredictable (PU) and Motor Only (MO). In the active conditions, the participants had full motor control over the stimulus onset, while in the passive conditions, they had no motor control. The stimulus was presented at a fixed time in the predictable conditions, while in the unpredictable conditions, the stimulus appeared randomly. MO condition consisted of button presses at given times, with no STM task related. Contrary to our previous results, our current results show that the AP condition performs worst (accuracy mean = 0.79). Two-way paired ANOVA yields a significant main effect of predictability on accuracy only (p = 0.004). Paired t-tests corrected by Holm method show a significant difference between AP and AU (p = 0.034), and between AP and PU (p = 0.043). These results suggest that STM is taxed more strongly when subjects control the onset of the stimulus and the time of presentation is predictable, but also indicate that predictability seems to be a major factor in this effect.