Do we know what we need? Preference for feedback about accurate performances does not benefit sensorimotor learning.

Abstract

Previous research on skill acquisition has shown that learners seem to prefer receiving knowledge of results (KR) about those trials in which they have performed more accurately. In the present study, we assessed whether this preference leads to an advantage in terms of skill acquisition, transfer, and retention of their capacity to extrapolate the motion of decelerating objects during periods of visual occlusion. Instead of questionnaires, we adopted a more direct approach to investigate learners' preferences for KR. Participants performed 90 trials of a motion extrapolation task (acquisition phase) in which, every three trials, they could decide between receiving KR about their best or worst performance. Retention and transfer tests were carried out 24 hr after the acquisition phase, without KR, to examine the effects of the self-selected KR on sensorimotor learning. Consistent with the current literature, a preference for receiving KR about the most accurate performance was observed. However, participants' preferences were not consistent throughout the experiment as less than 10% (N = 40) selected the same type of KR in all their choices. Importantly, although preferred by most participants, KR about accurate performances had detrimental effects on skill acquisition, suggesting that learners may not always choose the KR that will maximize their learning experiences and skill retention. (PsycINFO Database Record

On Whether Task Experience of the Peer Differentially Impacts Feedback Scheduling and Skill Acquisition of a Learner

“…The results from McRae et al (2015) showed that learners who received KR from a peer without task experience and those performers self-controlling their KR demonstrated similar learning. In addition to the motor performance similarities between the two conditions, McRae et al (2015) also reported that participants in the self-controlled KR condition preferred requesting KR after perceived good trials, consistent with previous self-controlled KR research (e.g., Chiviacowsky and Wulf, 2005; Bastos et al, 2018). Peers without task experience preferred to provide KR to the learner after both good and bad trials equally.…”

“…Curiously, the preference self-reported by the peers in the present experiment is also contrary to the preference self-reported by learners in past self-controlled feedback conditions. Previous self-controlled KR research has shown a preference for learners to request feedback more frequently after perceived good, rather than perceived poor trials (Chiviacowsky and Wulf, 2002, 2005; Wulf, 2007; Patterson and Carter, 2010; Hansen et al, 2011; c.f., Aiken et al, 2012; Laughlin et al, 2015; Bastos et al, 2018). This preference has been attributed to the decreased cognitive processing demands attributed to repeating the motor commands of a correct response, compared to engaging in the cognitive demands required to fix an error (Lam et al, 2010).…”

The effect of freedom of relevant choice on the 7-m throw in university students practicing handball

“…Studies have used artificial tasks such as sequential timing with computer keys (e.g. Aiken et al, 2020; Bastos et al, 2018; Chiviacowsky & Wulf, 2002; Keetch & Lee, 2007; Patterson & Carter, 2010; Wu & Magill, 2011) as well as real tasks such as golf putting (e.g. Jaquess et al, 2019; Nunes et al, 2018), crawl swim (e.g.…”

Estimativa de erro e conhecimento de resultados autocontrolado na aprendizagem motora

“…Foi requisitado que os participantes utilizassem o mouse com a mão dominante e a altura do monitor foi ajustada, de forma que o centro deste estivesse na altura dos olhos, a uma distância aproximada de 1 m (BASTOS et al, 2018). No início da prática, no caso do grupo Auto-com, e nos testes de RETi, RETa, TR, para ambos os grupos, foi requisitado que os participantes manifestassem verbalmente a estimativa de erro utilizando a palavra "antes", se avalia que apertou o botão antes do retângulo chegar na linha de contato; e utilizasse a palavra "depois" caso achasse que houve atraso na resposta.…”

“…Outra possível limitação pode ter sido o número de tentativas da fase de aquisição. Embora tenha se aumentado o número de tentativas quando comparado aos estudos que utilizaram uma tarefa semelhante de timing coincidente (90 tentativas) (DREWS, 2017;BASTOS, et al, 2018), um número maior de tentativas talvez possibilitasse verificar diferença na medida de precisão da estimativa de erro entre os blocos da aquisição. Adicionalmente, é possível supor que essa diferença fosse verificada, caso a oclusão do retângulo fosse maior durante toda a fase de aquisição.…”