Academic journal article Memory & Cognition

Exploring Specificity of Speeded Aiming Movements: Examining Different Measures of Transfer

Academic journal article Memory & Cognition

Exploring Specificity of Speeded Aiming Movements: Examining Different Measures of Transfer

Article excerpt

Participants were trained and tested to move a mouse cursor from a start position to targets on a circular display in a perceptual-motor reversal condition, with horizontal, but not vertical, reversals. During training, some participants (experimental) moved to two targets either along a single diagonal axis (D1) or along both axes (D2). For D2, return movements from the targets were in the same direction as instructed movements to unpracticed targets. Others (control) trained on all targets. Testing always involved all targets. At test, movement times (to reach the target after leaving the start position) were shorter on trained than on untrained targets, especially for the D1 condition, documenting training specificity. However, movement times in the experimental conditions to new targets during testing were shorter than those in the control condition during training, documenting transfer of learning, with more transfer for D2 than for D1. Initiation times (to leave the start position after target onset) showed no transfer. The results provide evidence that specificity and transfer are not mutually exclusive and depend on the measure used to assess performance.

Researchers interested in understanding the conditions that promote skill transfer have often demonstrated that learners lack the ability to generalize beyond the situations presented during training. These findings are consistent with various theoretical accounts, including those involving transfer-appropriate processing (Morris, Bransford, & Franks, 1977; Roediger, Weldon, & Challis, 1989), encoding specificity (Tulving & Thomson, 1973), procedural reinstatement (Healy, Wohldmann, & Bourne, 2005), and specificity of practice in motor learning (Proteau, Marteniuk, & Lévesque, 1992). One purpose of the present study is to further explore the conditions demonstrating specificity of learning and those promoting generalizability of learning.

There are ways to enhance transfer of knowledge, and one such way is by introducing sources of variability in practice (e.g., Schmidt & Bjork, 1992; Wulf & Schmidt, 1997). However, the advantages of training variability appear to be limited to certain conditions, and these conditions are not well understood. For example, in a previous study (Healy, Wohldmann, Sutton, & Bourne, 2006), we found no advantage for variability of practice. Specifically, participants learned to perform a speeded aiming task, which involved finding on the display and moving a mouse cursor from a center start position to target digits that appeared along the circumference of a circular stimulus array (see Figure 1). During training, the participants practiced various types of perceptual-motor reversals, including a horizontal reversal (in which movements of the mouse to the left resulted in cursor movements to the right, and vice versa, but vertical movements remained intact), a vertical reversal (in which vertical movements were reversed but horizontal movements remained intact), and both types of reversals in one condition. Practice on one reversal resulted in improved performance on that reversal during training, and learning persisted across a 1-week retention interval. However, the learning was highly specific and did not transfer to the use of a mouse with a different type of reversal. In addition, training with a variety of reversals did not result in more transfer than training with only the single reversal included during testing. Thus, the findings did not support the variability-of-practice hypothesis but, instead, were explained in terms of a global inhibition strategy, according to which all normal mouse movements are inhibited when participants encounter a reversed mouse. However, for a mouse with a reversal along only one dimension, participants must disinhibit movements in order to reach the targets along a nonreversed dimension. Because the disinhibition requires an additional step in processing, applying the global inhibition strategy is predicted to result in longer movement times to targets along the nonreversed dimension than to those along the reversed dimension-a pattern of movement times that was found by Healy et al. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.