Influences of Attentional Focus on Postural and...

Journal of Motor Behavior, 2004, Vol. 36, No. 2, 189-199

Gabriele WulfJohn MercerDepartment of KinesiologyUniversity of Nevada, Las VegasNancy McNevinDepartment of Physical TherapyWayne State Universityand Rehabilitation Institute of Michigan, Detroit

ABSTRACT. The authors examined the influence that attentionalfocus on either a postural or a suprapostural task had on the per-formance of each task. Participants (N = 32) stood on an inflatedrubber disk and held a pole horizontally. All participants performedunder 4 attentional focus conditions: external (disk) or internal(feet) focus on the postural task, and external (pole) or internal(hands) focus on the suprapostural task. Compared with internalfocuses, external focuses on either task resulted in similar andreduced postural sway. Response frequency on each task increasedwhen participants focused on the respective task. Finally, an exter-nal focus on either task produced higher frequencies of respondingon the suprapostural task. The authors conclude that supraposturaltask goals have a stronger influence on postural control than viceversa, reflecting the propensity of the motor system to optimizecontrol processes on the basis of the desired movement effect.

Key wvords: balance, focus of attention, motor control, suprapos-tural task

ne of the most fascinating features of the humanmotor control system is its ability to optimize the con-

trol of movements as a function of the performer's intentionor attention. An example of adaptations resulting fromchanges in the individual's intention is adjustments in pos-tural control on the basis of suprapostural task goals (e.g.,Balasubramaniam & Turvey, 2000; Riley, Stoffregen, Gro-cki, & Turvey, 1999; Stoffregen, Pagulayan, Bardy, & Het-tinger, 2000; Stoffregen, Smart, Bardy, & Pagulayan, 1999).Riley et al. (1999), for instance, found that relative to abaseline condition (i.e., not touching the curtain), posturalsway was reduced if individuals were instructed to lightlytouch a curtain and to try to minimize movements of thecurtain. In contrast, when touching the curtain was deemedirrelevant for the experiment, no reduction in postural swaywas found. Their interpretation was that the addition of asuprapostural goal (i.e., keeping the curtain still) resulted inspontaneous reductions in postural fluctuations to facilitate

Mark A. GuadagnoliDepartment of KinesiologyUniversity of Nevada, Las Vegas

the achievement of the suprapostural goal. Similarly, Stof-fregen et al. (2000) demonstrated that when standing par-ticipants were instructed to search for letters in a block oftext displayed in front of them, postural sway was reducedas compared with when they looked at a blank piece ofpaper. Therefore, the constraints imposed by the suprapos-tural task (or the performer's intention) appear to influencepostural adjustments, thus facilitating goal achievement(e.g., not moving the curtain, finding target letters).

In another series of studies, the individual's focus ofattention has also been found to affect the motor controlprocesses involved in the perfornance of motor skills,including those requiring postural control or balance (e.g.,Shea & Wulf, 1999; Wulf, HoB, & Prinz, 1998; Wulf,Lauterbach, & Toole, 1999; Wulf, McNevin, & Shea, 2001;

for a review, see Wulf & Prinz, 2001). Specifically, thoseinvestigators compared the effects of attention directedeither to the movements themselves (internal focus) or tothe effects of those movements on an apparatus or imple-ment (external focus). Some of the researchers used adynamic balance task that required participants to balanceon a stabilometer and to minimize deviations of the balanceplatform from the horizontal (e.g., Wulf et al., 1998, Exper-iment 2; Wulf, McNevin, & Shea, 2001; Wulf, Shea, &Park, 2001). The results consistently demonstrated that par-ticipants required to focus externally (i.e., on keeping mark-ers on the platform horizontal) produced more effective per-formance and learning than did participants required tofocus internally (i.e., on keeping their feet horizontal). Wulf

Correspondence address: Gabriele Wulf Department of Kinesi-ology, University of Nevada, Las Vegas, 4505 Maryland Park-vay,Las Vegas, NV 89154-3034, USA. E-mail address: [email protected]

189

r%neciprocal Influences of AttentionalFocus on Postural and SupraposturalTask Performance

G. Wulf, J. Mercer, N. McNevin, & M. A. Guadagnoli

and her colleagues (e.g., McNevin, Shea, & Wulf, 2003;Wulf, McNevin, & Shea, 2001; Wulf & Prinz, 2001) haveargued that the advantage of adopting an external focus ofattention is that it allows unconscious or automatic process-es to control the movements. In contrast, when participantsadopt an internal focus, they are more likely to consciouslyintervene in those control processes, thereby inadvertentlydisrupting automatic control processes.

In a couple of recent studies, it has been found that theattentional focus can also modify the effects of intentions orsuprapostural task goals. McNevin and Wulf (2002) used atask similar to that used by Riley et al. (1999) and comparedthe effects of internal versus external focus instructions.Riley and his colleagues' instructions were ambiguous froman attentional focus point of view in that participants wereinstructed to "minimize movements of the curtain thatmight result from their touch" (p. 805). Thus, it was notentirely clear whether the instructions directed participants'attention more to keeping their finger still (internal focus)so that they would minimize movements of the curtain ormore to keeping the curtain still (external focus). McNevinand Wulf specifically instructed participants to touch a

loosely hanging sheet with a fingertip and to focus either onminimizing the movements of the finger (internal) or onminimizing the movement of the curtain (external). Theirresults showed that although both focuses resulted in simi-lar amounts of postural sway, the frequency of movementadjustments was clearly higher under external than underinternal focus conditions. Because postural control wasenhanced via increased reactivity to fluctuations in posturalsway when participants' attention was directed to the effectsof their actions rather than to the actions themselves, theauthors interpreted those findings as further support for theadvantages of an external focus of attention. Thus, com-pared with an internal focus, an external focus on the supra-postural task resulted in greater postural stability-provid-ing preliminary evidence that the type of attentional focuson the suprapostural task can have a significant qualifyinginfluence on postural control.

In a subsequent study, Wulf, Weigelt, Poulter, andMcNevin (2003) used a more challenging postural task.Rather than standing still on a stable surface, participantswere required to learn to balance on a stabilometer. Wulf andcolleagues manipulated participants' focus of attention on asuprapostural task (i.e., holding a tube horizontal in front oftheir stomach) by instructing different groups of learners tofocus either on their hands (internal focus) or on the tube(external focus) while balancing on the stabilometer plat-form. Performances on both the suprapostural task and thebalance (postural) task were measured. The results showedthat not only did an external focus enhance performance onthe suprapostural task (i.e., the stability of the tube), it alsoenhanced performance on the balance task (i.e., the stabilityof the platform). Even more interesting, the superior balanceperformance demonstrated by the external focus group incomparison with the internal focus group was even seen

when the suprapostural task was removed in a transfer test.Thus, compared with an internal focus, practice with anexternal focus on the suprapostural task had a lasting effect:It enhanced the learning of the balance task, for which nodifferential instructions had been given.

The findings from previous studies have demonstrated thatthe type of attentional focus on suprapostural tasks is impor-tant and as such needs to be considered when exarniningeffects of suprapostural tasks. As compared with the effect ofinternal focus, not only can an external focus on the supra-postural task enhance performance on the suprapostural task(Wulf et al., 2003), it can also result in more effective postur-al control (McNevin & Wulf, 2002; Wulf et al., in press). Aninteresting question is how the addition of a supraposturaltask functions to influence postural control and, more specif-ically, how an external focus of attention acts to optimize pos-tural stability. That is one of the questions addressed in thepresent study. In contrast to previous studies in which supra-postural task performance was not measured (e.g., McNevin& Wulf, 2002; Riley et al., 1999) or a crude performancemeasure was used (Wulf et al., 2003), we measured perfor-mance on both postural and suprapostural tasks in terms of

performance stability (root mean square error [RMNSE) andfrequency of responding (mean power frequency [MPF]).Whereas RMSE is an overall performance measure that indi-cates the average deviations from a goal (e.g., the horizontal),MPF is seen as an indicator of the control mechanisms usedto perform a task (e.g., McNevin et al., 2003; Newell &Slifkin, 1996; Wulf, McNevin, & Shea, 2001; Wulf, Shea, &Park, 2001). Specifically, a higher frequency of movementadjustments is viewed as a result of a more automatic, reflex-type mode of control that is based on faster and more finelytuned integrated movement responses. Balance on an unsta-ble surface involves continuous interplay between sensoryinput and effector output: The greater the coherence betweensensory input and effector output (McAuley & Marsden,2000), the finer the graded adjustments in postural control. Ifthe intention is to minimize postural sway, then smaller oscil-lations and higher frequency of responding would achieve thegoal of maintaining a stable base of support. Performancemediated by a high MPF suggests a system capable of adjust-ing to minor, as well as major, perturbations to the base ofsupport. In contrast, a low frequency of responding isassumed to reflect less coherence between sensory input andeffector output. The absence of confluence may be caused bya small, but functionally important, disruption between thesensory consequences of an action and effector output as adirect result of participants' consciously intervening in thoseprocesses.

In the present study, we wanted to examine to what extentthe attentional focus on the suprapostural task affects theoverall stability of the postural system (RMSE) and theautomaticity of the postural responses (MPF). ReducedRMSE (Wulf et al., 2003) or increased MPF (McNevin &Wulf, 2002) on the postural task has been found in previousstudies when the attentional focus on the suprapostural task

Journal of Motor Behavior190

Attentional Focus on Postural and Suprapostural Tasks

was external rather than internal. The apparent relationshipbetween the outcome (postural sway) and the control strat-egy (frequency of responding) used under the external focuscondition (especially in the McNevin & Wulf study)seemed to suggest that individuals maintain stability byincreasing their reactivity to perturbations. That is, whenparticipants used an external focus, they appeared to com-pensate for an undesired outcome (increased postural sway)by increasing their responsiveness (MPF) to postural devia-tions. In contrast, the internal focus condition resulted in alower frequency of responding, suggesting that participants

were less effective in compensating for increases in postur-al sway. In the present study, we wanted to follow up onthose findings. In particular, we wanted to further examinewhether those two performance characteristics are related.For example, would reduced postural sway (i.e., reducedRMSE) be accompanied by an increase in frequency of pos-tural adjustments (i.e., higher MPF)? In contrast, if the twoperformance characteristics are not related, would an exter-nal focus lead to reduced sway without associated increas-es in response frequency, or vice versa?

Furthermore, we asked whether the type of focus adopt-ed on the postural task would affect performance on the

suprapostural task. If, for example, an external focus on apostural task affected the whole motor system (cf. McNevin& Wulf, 2002), then its beneficial effects might generalizeto performance on the suprapostural task. In contrast, ifattention is directed (solely) to the postural task, then thesuprapostural task might become more or less irrelevant andthe effects of attentional focus, if any, might be minimal. Atany rate, it seerned important to examine reciprocal influ-ences of attentional focuses on suprapostural and posturaltasks in order to gain a better understanding of how inten-tion and attention influence the motor control system and,

as a consequence, an individual's performance.We used a within-participant design to address those

questions. Under all conditions, participants were requiredto stand on an inflated rubber disk. We used the disk to makethe postural task more challenging in comparison with acondition that required the participant to stand on a flat andstable surface of support (McNevin & Wulf, 2002; Riley etal., 1999). In addition, participants were asked to hold a polehorizontally in front of their stomach (suprapostural task;see Figure 1). All participants performed under each of fourconditions: internal or external focus on the postural taskand internal or external focus on the suprapostural task. We

June 2004, Vol. 36, No. 2

FIGURE 1. Schematic of a participant balancing on the disk and holding the pole.

191


measured performances on both postural and supraposturaltasks to determine the effects of the type of attentional focuson either task as well as possible interaction effects.

It should be pointed out that in the present study, weviewed reduced sway as an indication of enhanced posturalperformance. That is, less postural sway was equated withgreater stability and thus better performance. That point isimportant, because Riley et al. (1999), Stoffregen et al.(2000), and Stoffregen et al. (1999) have rightly empha-sized that postural stability needs to be seen in the contextof the suprapostural goal. In fact, in certain situationsincreased postural sway may be more adaptive than lesssway in that it facilitates the achievement of the suprapos-tural goal (e.g., when tracking a visual target). In the presentstudy, however, the suprapostural goal of holding the wandhorizontally should be facilitated by reduced postural sway.

Method

Participants

Thirty-two university students participated in this exper-iment. None had prior experience with the task, and all werenaive as to our purpose in the experiment. All participantsgave their informed consent. They received extra coursecredit for their participation.

Apparatits and Task

In the postural task; we required participants to balanceon a 13-in.-diameter inflated rubber disk (Disc 'O' Sit; Per-form Better, Cranston, RI). One often uses disks such asthose in athletic training to strengthen the muscles involvedin balance. The disk, which was placed on a force plate(Kistler Corp., Amherst, NY, Model #9286AA), recordedcenter of pressure data. While standing on the disk, partici-pants held a round PVC pole, 195 cm in length and 1.6 cmin diameter (see Figure 1), on which a uniaxial accelerom-eter (PCB Piezotronics; Depew, NY, Model #478A01) wasmounted so that the sensitive axis was aligned vertically.We used the accelerometer to record information regardingthe vertical sway of the pole. Units for acceleration weremultiples of the acceleration caused by the force of gravity.We recorded center of pressure and acceleration data con-currently at 500 Hz during each trial. Participants wereinstructed to grasp the pole, with their palms facing upward,

and to put their rmiddle fingers on yellow, 1-cm-wide stripesof tape that surrounded the pole at a distance of 19 cm to theleft and right from the center of the pole. Participants wereasked to hold the pole horizontally, with their elbows flexedat 90°. We instructed them to look straight ahead (i.e., notto look at the pole). Each participant performed three 15-strials under each of four conditions (see the following).

Procedutre

Participants were informed that the task was to stand on thedisk and hold the pole horizontally. All participants per-formed three 15-s trials under each of four attentional focusconditions. We designed the instructions given in those con-

ditions so that they would induce an internal or external focuson the postural or suprapostural task. Specifically, weinstructed participants to focus on minimizing the movementsof their feet (postural task, internal focus), minimizing themovements of the disk (postural task, external focus), holdingtheir hands still (suprapostural task, internal focus), or hold-ing the pole still (suprapostural task, external focus). Theorder of conditions was counterbalanced across participants.

Dependent Variables and Data Analysis

We converted the data to ASCII format and processedthem with custom laboratory software. We adjusted centerof pressure (COP) data so that the central coordinates were(0,0). Data were then converted from Cartesian to polarcoordinates. We analyzed the magnitude vector by calculat-ing the root mean square error (RMSE) and mean power fre-quency (MPF) of the signal. We calculated MPF after firsttransforming the vector magnitude signal to the frequencydomain by calculating the power spectral density (PSD) ofthe signal. We calculated the PSD with the mean and lineartrends removed from the signal. Because the length of thedata set must be a power of 2, we padded the 7,500 datapoints (15 s of data recorded at 500 Hz) with zeros to obtain8,192 points for use in calculating the PSD. That procedureresulted in 0.061-Hz frequency bin sizes. MPF was calcu-lated as Y(PSDi * Freqi)/X(PSD), where PSDi representspower spectral density at frequency bin i, Freq1 representsthe frequency (in Hz), and Y(PSD) represents the sum ofpowers across all frequency bins analyzed (Winter & Patla,1997). The 0- to 20-Hz frequency range was used becausewe had calculated that 97-99% of the power was containedwithin that range. RMSE of the COP vector magnitudeserved as a measure of postural sway, whereas MPF indi-cated the frequency of postural adjustments.

We calculated the RMSE of the acceleration signal aftersmoothing the data with a fourth-order, low-pass filter withthe optimal cut-off frequency identified (Winter & Patla,1997). We calculated MPF from the raw acceleration signal,using the identical procedures described for COP data. TheRMSE of the (vertical) acceleration of the pole was used asa measure of the overall stability of the pole, whereas theMPF of the acceleration signal was a measure of theresponse frequency on the suprapostural task. We analyzedeach dependent variable in a 2 (task: postural, suprapostur-

al) x 2 (focus of attention: internal, external) x 3 (trial)repeated measures analysis of variance.

Results

Postural Task: Magnitude of Sway (RMSE)

External focus instructions generally resulted in less pos-tural sway (i.e., smaller RMSE) than internal focus instruc-tions did (see Figure 2). That is, independent of whetherattention was directed at the postural or at the suprapostur-al task, postural stability was greater when participantsfocused on the movement effect (disk or pole) than whenthey focused on the body movements producing the effect



(feet or hands). In fact, external and internal focuses on thesuprapostural task produced RMSE values that were numer-ically very similar to those produced by the respectivefocuses on the postural task. The main effect of focus ofattention was significant, F(1, 31) = 4.18, p < .05, co2 = .91,whereas the main effects of task, F(1, 31) < 1, and trial, F(2,62) = 1.58, p > .05, were not significant. Also, none of theinteraction effects was significant, all Fs < 1.

Postural Task: Frequency of Responding (MPF)

As can be seen in Figure 3, the frequency of respondingwas clearly greater when participants focused on the pos-tural task than when they focused on the supraposturaltask. That is, independent of the type of attentional focus(internal or external), postural adjustments occurred at ahigher rate when individuals directed their attention to thepostural task. The task main effect was significant, F(1,31) = 6.62, p < .05, 0)2 = .15, whereas the main effects offocus of attention, F(1, 31) < 1, and trial, F(2, 62) = 1.7,p > .05, were not significant. No interaction effects weresignificant.

Suprapostural Task: Magnitude of Sway (RMSE)

The amount of sway observed for the pole did not differas a function of task or attentional focus (see Figure 4). Thatis, at least with respect to that measure of suprapostural taskperformance, performances were similar-regardless ofwhether participants focused on the suprapostural or thepostural task or whether they adopted an internal or exter-

nal focus. The main effects of task, F(1, 31) = 2.48,p> .05,focus of attention, F(1, 31) < 1, and trial, F(2, 62) < 1, werenot significant. Also, none of the interaction effects was sta-tistically significant.

Suprapostural Task: Frequency of Responding (MPF)

With regard to the frequency of responding, there wereclear performance differences as a function of both task andattentional focus (see Figure 5). The frequency of move-ment adjustments was considerably higher when partici-

pants focused on the suprapostural task than when theyfocused on the postural task. Also, higher movement fre-quencies were seen when participants adopted an externalfocus (on either task) than when they adopted an internalfocus. The main effects of both task, F(1, 31) = 64.83, p <.001, d = .67, and focus of attention, F(1, 31) = 6.75, p <.05, ()2 = .15, were significant. Even though the effects ofattentional focus appeared to be larger for the suprapostur-al than for the postural task, the interaction of task andattentional focus failed to reach significance, F(1, 31) =2.90, p = .10. Also, the trial main effect, F(2, 62) < 1, andthe other interaction effects were not significant. Thus, bothfocusing on the suprapostural task and adopting an externalfocus resulted in increased response frequencies on thesuprapostural task.

Discussion

In the present study, we were motivated by recent find-ings demonstrating the effects of a suprapostural task on

June 2004, Vol. 36, No. 2

El Internal

E External

1.6 Feet Disk Hands Wand

1.5

1.4-

1.3-

E 1.2 -

1.0

0.9

0.8-

0.7-

0.6-Postural task Suprapostural task

FIGURE 2. Magnitude of sway (root mean square error, RMSE) on the postural task as afunction of the task to which attention was directed (postural or suprapostural task) and thetype of attentional focus adopted (external or internal).

193


Feet Disk

T

T

El Internal

U External

0.10 -

0.09 -

0.08 -

0.07 -

0.06 -

0.05 -

0.04 -

0.03 -

Hands Wand

0.02

0.01

0.00Postural task Suprapostural task

FIGURE 4. Magnitude of sway (root mean square error, RMSE) on the suprapostural task asa function of the task to which attention was directed (postural or suprapostural task) and thetype of attentional focus adopted (external or internal).

postural stability (e.g., Riley et al., 1999; Stoffregen et al.,2000), as well as by findings demonstrating that an individ-

ual's focus of attention on the suprapostural task can play animportant modifying role in that context (McNevin & Wulf,2002; Wulf et al., 2003). The latter studies showed that pos-

tural stability or balance performance (and learning) wasenhanced only when participants were instructed to adopt

an external focus of attention on the suprapostural task, butnot when they adopted an internal focus (relative to noinstructions). One of our goals in the present study was to

Journal of Motor Behavior

| Internal

U1 External

0.54 Feet Disk Hands Wand

0.52 -

N 0.50 -

U-a. 0.48

0.46


FIGURE 3. Frequency of responding (mean power frequency, MPF) on the postural task asa function of the task to which attention was directed (postural or suprapostumal task) and thetype of attentional focus adopted (external or internal).

0

LUco)

194


provide further insight into how the attentional focus on thesuprapostural task acts to affect postural control. We usedpostural sway and response frequency measures similar tothose used by McNevin and Wulf (2002). However, todetermine the generalizability of their results, we used amore difficult postural task than McNevin and Wulf had

used. In addition to examining the influence that the type ofattentional focus on the suprapostural task has on posturalcontrol, we wanted to investigate the degree to which thefocus of attention on the postural task would affect perfor-mance of the suprapostural task. One drawback of previousstudies is that performance on the suprapostural task wasnot measured at all (e.g., McNevin & Wulf, 2002; Riley etal., 1999) or that a relatively crude measure was used (Wulfet al., 2003). Therefore, in the present study, we used moresensitive measures of suprapostural task performance. Wesurmised that, although suprapostural task effects on pos-tural control appear to be comparatively strong, the reverseeffect, if any, would be weaker. The explanation for that dif-ference in effect is that postural control subserves the supra-postural task goal. That is, suprapostural performancedepends on the degree of stability exerted over the posturaltask. However, focusing on the postural task (internally orexternally) might have relatively little consequences forsuprapostural task performance. In the following sections,we discuss, in tum, influences on postural task perfor-mance, influences on suprapostural task performance, andreciprocal influences of suprapostural and postural tasks asa function of attentional focus.

Attentional-Focus Influenceson Postural Task Performance

In line with previous findings (e.g., McNevin et al., 2002;Shea & Wulf, 1999; Wulf et al., 1998; Wulf, McNevin, &Shea, 2001), instructions to adopt an external focus on thepostural task (i.e., to focus on minimizing movements of the

disk) resulted in less postural sway than did instructions toadopt an internal focus (i.e., to focus on minimizing move-ments of the feet). The stabilometer task was used in previ-ous studies, and participants were required to balance on theplatform in accordance with external (focus on markersattached to the platform) or internal (focus on feet) focusinstructions. Performance was measured in terms of theaverage deviation from the horizontal. In the present study,participants were required to balance on an inflated diskthat allowed deviations in all directions and required corre-sponding movement adjustments. The results, however,were similar in showing external focus advantages. Thoseresults demonstrate that the external focus benefits are gen-eralizable to various balance tasks (see also Landers, Wulf,Wallmann, & Guadagnoli, 2003; Totsika & Wulf, in press).

Important in the present context was the finding that,compared with the respective internal focuses, postural sta-bility was enhanced not only by an external focus on thepostural task but also by an external focus on the suprapos-tural task. The beneficial effect on postural control as aresult of an external focus on the suprapostural task is inline with the finding of Wulf et al. (2003), who had partici-pants hold a tube horizontal while balancing on the sta-

June 2004, Vol. 36, No. 2

o Internal

E External

5.9 - Feet Disk Hands Wand

5.7 -

5.5-N

1 5.3-U-a.

5.1

4.9-

4.7-


FIGURE 5. Frequency of responding (mean power frequency, MPF) on the supraposturaltask as a function of the task to which attention was directed (postural or suprapostural task)and the type of attentional focus adopted (external or internal).

195


bilometer. They found that balance performance (and learn-ing) was enhanced when participants adopted an externalfocus on the suprapostural task (i.e., focused on the tube) incomparison with their performance when they adopted aninternal focus (i.e., focused on their hands). Thus, theresults of Wulf et al. and the present study are consistent inshowing that the type of attentional focus on the suprapos-tural task affects postural task performance and that anexternal focus leads to greater postural stability. McNevinand Wulf (2002) came to the same conclusion. Even thoughthey did not find differences between external and internalfocuses in terms of postural sway, participants' responsefrequency (MPF) was greater under external than underinternal focus conditions. McNevin and Wulf's interpreta-tion was that postural stability is optimized under externalconditions as a result of greater frequency in responding,which compensates for increased postural sway viaenhanced reactivity to perturbations.)

Most interesting, external versus internal focuses on thepostural or suprapostural tasks resulted in very similaramounts of postural sway, respectively. That is, an externalfocus led to about the same magnitude of sway, independentof whether the focus was on the postural task (disk) or thesuprapostural task (pole). Similarly, an internal focus pro-duced almost identical degrees of sway, independent ofwhether attention was directed to the postural task (feet) orthe suprapostural task (hands). Based on those findings, itmight appear that postural stability is generally enhancedby an external focus and that the exact locus of attention isirrelevant. However, the response frequency results sug-gested otherwise.

Participants showed a clearly higher response frequency(MPF) in postural control when their attention was directedto the postural task-independent of the type of attentionalfocus-than when their attention was directed to the supra-postural task. That is, whereas external and internal focusesyielded similar MPF values on each task, focusing on thepostural task led to higher MPFs than did focusing on theother (suprapostural) task. The absence of an attentionalfocus main effect (i.e., higher MPFs for external relative tointernal focus conditions) is at odds with the findings ofprevious studies (e.g., McNevin & Wulf, 2002; Wulf,McNevin, & Shea, 2001; Wulf, Shea, & Park, 2001). How-ever, when combined with the postural sway (RMSE)results, it becomes clear that an external focus on the pos-tural task resulted in greater postural stability than did anyother condition: Focusing on the disk produced a relativelysmall amount of postural sway and a high response fre-quency. Thus, not only did participants show reduced swaywhen they focused on the disk, they were also able to react

faster to a loss of balance because of the increased frequen-cy of responding.

Conversely, an internal focus on the suprapostural task(hands) resulted in the least effective postural performance,with a relatively large amount of postural sway and a lowresponse frequency. It is possible that the detriments in pos-

tural performance resulted from the higher attentionaldemands associated with an internal than with an externalfocus (Wulf, McNevin, & Shea, 2001). The results of stud-ies examining the attentional demands of postural controlhave shown that the attentional resources required to main-tain balance increase as the balance task becomes more chal-lenging (or if the motor control system is compromised, e.g.,by age or disease; for a review, see Woollacott & Shumway-Cook, 2002). Standing on an unstable surface such as theinflated disk used in the present study can therefore beassumed to require at least a certain amount of attentionalresources. By adding an attention-demanding dual (supra-postural) task, some of those resources are presumably takenaway-resulting in degraded postural performance (e.g.,Shumway-Cook, Woollacott, Baldwin, & Kerns, 1997).Thus, it is possible that postural control suffered because theinternal focus on the suprapostural task required a relativelylarge amount of the attentional capacity.

It is also interesting to note that even though an externalfocus on the suprapostural task (pole) resulted in reducedpostural sway, similar to an external focus on the posturaltask (disk), focusing on the pole did not result in theincrease in response frequency seen when participantsfocused on the disk. Thus, even though postural stabilitywas enhanced when attention was directed to the pole, therewas no concomitant increase in the degree of automaticityof postural control, as was seen when attention was direct-ed to the disk. Apparently, in order for those benefits of anexternal focus to be manifested, one must direct attention tothe task at hand (in this case, the postural task).

In most previous studies in which investigators haveexamined response frequency as a function of attentionalfocus (e.g., McNevin et al., 2003; Wulf, McNevin, & Shea,2001; Wulf, Shea, & Park, 2001) the stabilometer task hasbeen used, and attention was directed either to the per-former's feet (internal) or to markers on the balance platform(external). In those studies, participants who adopted anexternal focus showed consistently higher MPFs than didthose who adopted an internal focus. In the present study,although there was a small tendency for the external focuscondition (disk) to produce higher MPFs than the internalfocus condition (feet) did, the overriding effect was pro-duced by the task to which the attention was directed (pos-tural or suprapostural). That result suggests that the ability torespond quickly to a perturbation is generally enhanced ifthe focus is on the postural task rather than on the suprapos-tural task. Nevertheless, an external focus on the posturaltask proved to be most effective because it resulted in highresponse frequencies in combination with reduced sway.

Attentional-Focus Influenceson Suprapostural Task Performance

There were no influences on the sway of the pole (RMSE)as a function of the type of task that was attended to or thetype of attentional focus. That is, regardless of whether par-ticipants focused on the suprapostural or postural task and



whether the focus was internal or external, RMSE valueswere similar. In contrast, Wulf et al. (2003) found superiorsuprapostural task performance (number of times a ballmade contact with either side of a tube that was supposed tobe held horizontally) for participants who adopted an exter-nal focus on the suprapostural task, in comparison withthose who adopted an internal focus. A possible reason forthe fact that no effects were found for RMSE in the presentstudy could be related to the nature of the task, which waspresumably less challenging than the one used by Wulf etal. In that study, participants were required to hold an objecthorizontal despite relatively large excursions of the balanceplatforn on which they were standing. In the present study,postural fluctuations were not as extensive, which presum-ably also facilitated the suprapostural task. It is possible thatbecause of that stability, no attentional focus effects wereseen in the amount of sway.

Nevertheless, the measure of response frequency on thesuprapostural task (MPF) did show significant effects ofboth the task attended to and the attentional focus. Focusingon the suprapostural task yielded clearly higher MPFs onthat task than did focusing on the postural task. The effectwas similar to that seen for postural task performance; inthat case, focusing on the postural task resulted in higherMPFs. Thus, it appears that there is a task-specific effect, inthat performance is enhanced on the task (suprapostural orpostural) to which attention is directed.

Most important, an external focus also resulted inincreased MPF values. The increase was independent ofwhether the focus was directed at the suprapostural or thepostural task. Even though it appeared that an external focusproduced even higher MPFs when participants focused onthe suprapostural rather than the postural task, the interac-tion of task and attentional focus was not significant. Theincreased response frequency seen under external relative tointernal focus conditions is in line with results of previousstudies (McNevin et al., 2003; Wulf, McNevin, & Shea,2001; Wulf, Shea, & Park, 2001) in which, in comparisonwith an internal focus, an external focus resulted in increas-es in response frequency on postural tasks. The presentresult shows that that effect generalizes to tasks other thanpostural tasks. It also provides additional support for theconstrained-action hypothesis (e.g., McNevin et al., 2003;Wulf, McNevin, & Shea, 2001), according to which, com-pared with an internal focus, an external focus promotes theuse of more automatic control processes. Those rapid,

reflex-like processes are assumed to be responsible for thefast movement adjustments observed under external focusconditions. Despite the main effect of attentional focus,clearly the largest numerical MPF values were obtainedwhen an external focus was adopted on the supraposturaltask (focus on pole). Given that the amount of sway on thesuprapostural task was similar for all conditions, the MPFresults suggest that the adoption of an external focus on thesuprapostural task is optimal for the performance of thattask. That conclusion concurs with that drawn from the pos-

tural performance results, in which an external focus wasalso found to be optimal.

Reciprocal Influences of Attentional Focuson Suprapostural and Postural Task Performance

The type of focus on either task had an effect on the per-formance of the other task. However, the influence of the typeof focus on the suprapostural task on postural task perfor-mance appeared to be stronger than vice versa. Focusing onthe pole, that is, adopting an external focus on the suprapos-tural task, led to the same (reduced) amount of postural swayas did focusing on the disk, that is, adopting an external focuson the postural task. Conversely, an internal focus on eithertask produced similar sway magnitudes. Thus, the type ofattentional focus on the suprapostural task clearly had aneffect on postural control. In contrast, the type of focus on thepostural task had only a relatively small effect on suprapos-tural task performance. Compared with focusing on the feet(internal focus), focusing on the disk (external focus) result-ed in a numerically small increase in response frequency onthe suprapostural task. Also, there was no effect on theamount of sway of the pole. Thus, whether individualsfocused externally or internally on the postural task had onlya small, if any, effect on suprapostural task performance.

How can one explain the imbalance in the reciprocalinfluence that the attentional focus on either task had on theperformance of the other task? As suggested earlier, supra-postural task performance is directly influenced by thedegree of stability of the supporting system (posture). Thatis, the act of balancing subserves the achievement of thesuprapostural task goal. For example, if the goal is to notspill a cup of coffee while traversing a hall, success in thataction will depend on how stable one's gait pattern is. Thus,one would expect to see influences on postural performanceas a function of the suprapostural task, and evidence for thatinfluence has come from a number of previous studies (e.g.,Balasubramaniam & Turvey, 2000; Riley et al., 1999; Stof-fregen et al, 2000; Stoffregen et al., 1999). The findings ofthe present study, together with those of McNevin and Wulf(2002) and Wulf et al. (2003), show that the type of atten-tional focus plays an important qualifying role in that con-text-an external focus on the suprapostural task typicallyleads to more effective postural control. It is possible that,compared with an internal focus, an external focus directedat the suprapostural task goal enhances postural stabilitybecause focusing on the outcome of the action tweaks the

control mechanisms that enable the individual to achievesuprapostural and postural stability.

In contrast, the type of focus on the postural task had arelatively minor effect on the performance of the suprapos-tural task. Even though focusing on the disk (external focus)produced clearly greater postural stability than did focusingon the feet (internal focus), the stability was not translatedinto greater stability of the pole (RMSE). (The only effectwas a small increase in response frequency.) Yet, the lack oftransfer can presumably be explained by the absence of a

June 2004, Vol. 36, No. 2 197


suprapostural task goal. Even though participants were stillholding the pole, their attention was directed to the act ofstanding still. Thus, the task of holding the pole horizontalbecame, at best, secondary, or perhaps even irrelevant (as inRiley et al., 1999). Therefore, it is perhaps not too surpris-ing that the different attentional focuses on the primary task(i.e., standing still) did not have much of an effect on thestability of the pole. That situation would be comparable toa condition in which an individual focuses on traversing thehall while carrying a cup of coffee. The attentional focus onthe postural (locomotor) task would not necessarily deter-mine how successfully one is able to avoid spilling the cof-fee. Thus, it appears that postural stability does not translateinto stability of other, functionally unrelated body sys-tems-if the primary goal is to maintain balance. Suprapos-tural task performance would be influenced only if the indi-vidual imposed constraints over the degrees of freedomhigher up in the link (i.e., in the upper body). Yet, if thesuprapostural goal is more or less irrelevant, then there is nocompelling reason to do so. The ability to dissociate func-tionally relevant from functionally irrevelant task goals canperhaps be seen as reflecting economy of the motor system,whereby control effort is invested only to the extent that itis necessary for the accomplishment of the goal.

Despite the reciprocal effects of the two tasks-in partic-ular the influence of the suprapostural task on postural con-trol-there were also clear task-specific effects. Focusingon the postural task resulted in higher response frequencieson the postural task, and-independent of the type of atten-tional focus-focusing on the suprapostural task resulted inhigher response frequencies on that task. Because motor

coordination is a function of coherence between sensoryinput and effector output (Marsden, Ashby, Limousin-Dowsey, Rothwell, & Brown, 2000), one could argue thatfocusing on the effector output of a specific activity dictateswhich sensory information is used, in turn, to organize theentire effector system (e.g., McAuley & Marsden, 2000). Assuch, the higher response frequencies associated with theparticular task focused on may reflect the tendency of thesystem to organize itself around what the task goal is (min-imize pole movement or minimize postural sway).

Summary and Conclusions

The present findings provide support for both the conceptof suprapostural goals (e.g., Balasubramaniam & Turvey,2000; Riley et al., 1999; Stoffregen et al, 2000; Stoffregen etal., 1999) and the importance of the performer's focus ofattention (for a review, see Wulf & Prinz, 2001). Specifical-ly, if the suprapostural task was made salient, then instruc-tions to focus on the pole (external focus) resulted in greaterpostural stability than did instructions to focus on the hands(internal focus). However, if attention was directed to thepostural task-making the suprapostural goal less salient -then the attentional focus on the postural task had only asmall effect, if any, on the performance of the supraposturaltask. We argue that the reason for the asymmetric effect is

that postural control subserves the suprapostural task goal,whereas the reverse is not necessarily the case (even thoughgreater instability of suprapostural task performance wouldalso affect postural stability). Most important, the type ofattentional focus had a significant qualifying effect: Anexternal focus generally resulted in more effective postural(and suprapostural) task performance than an internal focusdid (see also McNevin & Wulf, 2002; Wulf et al., 2003).Those benefits of an external focus of attention are in linewith the results of numerous previous studies that havedemonstrated more effective performance and learningwhen individuals adopt an external focus than when theyadopt an internal focus (Wulf & Prinz, 2001). Finally, thepresent results indicate that performance on a specific task(postural or suprapostural) is enhanced if attention is direct-ed to that task. That is, performance on the postural task wasmost effective if attention was directed to the postural taskand if the focus was external (disk). Similarly, performanceon the suprapostural task was best if an external focus ofattention was adopted on the suprapostural task (pole). Thus,despite suprapostural goal influences on postural control, thetask that was attended to (postural or suprapostural) was per-formed best. Overall, the present results confirm the view ofa "smart" motor system that optimizes the control processesbased on the environmental outcome, or movement effect,that the performer wants to achieve.

NOTE1. McNevin and Wulf (2002) found that in comparison with

when participants adopted an internal focus, when they adopted anexternal focus on the suprapostural task they showed an increasein response frequency (MPF), but there was no difference in pos-tural sway. In the present study, an external focus resulted inreduced postural sway (RMSE), but there were no differences inMPF. The discrepancy in the results may have been caused by dif-ferences between the postural (and perhaps suprapostural) tasks inthe two studies. In the McNevin and Wulf study, participants stoodon a stable surface, with their eyes closed; in the present study,however, participants balanced, with their eyes open, on an unsta-ble surface. In future studies, investigators should more systemat-ically examine how task features such as the stability of the sur-face or the availability of visual feedback affect various aspects ofpostural performance.

REFERENCES

Balasubramaniam, R., & Turvey, M. T. (2000). The handedness ofpostural fluctuations. Human Movement Science, 19, 667-684.

Landers, M., Wulf, G., Wallmann, H., & Guadagnoli, M. A.(2003). An external focus of attention attenuates balanceimpairment in Parkinson's disease. Manuscript submitted forpublication.

Marsden, J., Ashby, P., Limousin-Dowsey, P., Rothwell, J., &Brown, P.. (2000). Coherence between cerebellar thalamus, cor-tex and muscles in man. Brain, 123, 1459-1470.

McAuley, J., & Marsden, C. (2000). Physiological and pathologi-cal tremors and rhythmic central motor control. Brain, 123,1545-1567.

McNevin, N. H., Shea, C. H., & Wulf, G. (2003). Increasing thedistance of an external focus of attention enhances learning.Psychological Research, 67, 22-29.



McNevin, N. H., & Wulf, G. (2002). Attentional focus on supra-postural tasks affects postural control. Human Movement Sci-ence, 21, 187-202.

Newell, K. M., & Slifkin, A. B. (1996). The nature of movementvariability. In J. Piek (Ed.), Motor control and human skill: Amultidisciplinary perspective. Champaign, IL: Human Kinetics.

Riley, M. A., Stoffregen, T. A., Grocki, M. J., & Turvey, M. T.(1999). Postural stabilization for the control of touching.Human Movement Science, 18, 795-817.

Shea, C. H., & Wulf, G. (1999). Enhancing motor learning throughexternal-focus instructions and feedback. Human MovementScience, 18, 553-571.

Shumway-Cook, A., & Woollacott, M., Baldwin, M., & Kerns, K.(1997). The effects of cognitive demands on postural sway inelderly fallers and non-fallers. Journal of Gerontology: MedicalSciences, 52, M232-M240.

Stoffregen, T. A., Pagulayan, R. J., Bardy, B. G., & Hettinger, L. J.(2000). Modulating postural control to facilitate visual perfor-mance. Human Movement Science, 19, 203-220.

Stoffregen, T. A., Smart, L. J., Bardy, B. G., & Pagulayan, R. J.(1999). Postural stabilization of looking. Journal of Experimen-tal Psychology: Human Perception and Performance, 25,1641-1658. 1

Totsika, V., & Wulf, G. (in press). An external focus of attentionenhances transfer to novel situations and skills. Research Quar-terly for Exercise and Sport.

Winter, D. A., & Patla, A. E. (1997). Signal processing and linear

systems for the movement sciences. Waterloo, Ontario, Canada:University of Waterloo Press.

Woollacott, M., & Shumway-Cook, A. (2002). Attention and thecontrol of posture and gait: A review of an emerging area ofresearch. Gait and Posture, 16, 1-14.

Wulf, G., H1oB3, M., & Prinz, W. (1998). Instructions for motorlearning: Differential effects of internal versus external focus ofattention. Journal of Motor Behavior, 30, 169-179.

Wulf, G., Lauterbach, B., & Toole, T. (1999). Learning advantagesof an external focus of attention in golf. Research QuarterlyforExercise and Sport, 70, 120-126.

Wulf, G., McNevin, N. H., & Shea, C. H. (2001). The automaticityof complex motor skill learning as a function of attentional focus.Quarterly Journal of Experimental Psychiology, 54A, 1143-1154.

Wulf, G., & Prinz, W. (2001). Directing attention to movementeffects enhances learning: A review. Psychonomic Bulletin &Reviewv, 8, 648-660.

Wulf, G., Shea, C. H., & Park, J.-H. (2001). Attention in motorlearning: Preferences for and advantages of an external focus.Research Quarterly for Exercise and Sport, 72, 335-344.

Wulf, G., Weigelt, M., Poulter, D. R., & McNevin, N. H. (2003).Attentional focus on supra-postural tasks affects balance learning.Quarterly Journal of Experimental Psychology, 86, 1191-1211.

Submitted February 6, 2003Revised May 4, 2003

June 2004, Vol. 36, No. 2 199

COPYRIGHT INFORMATION

TITLE: Reciprocal Influences of Attentional Focus on Posturaland Suprapostural Task Performance

SOURCE: J Mot Behav 36 no2 Je 2004WN: 0415707509006

The magazine publisher is the copyright holder of this article and itis reproduced with permission. Further reproduction of this article inviolation of the copyright is prohibited.

Copyright 1982-2004 The H.W. Wilson Company. All rights reserved.

Influences of Attentional Focus on Postural and...

Documents

Transcript of Influences of Attentional Focus on Postural and...