Visual attentional preservation 1 Preservation of

Visual attentional preservation 1 Preservation of visual attention in older expert orienteers at rest and under physical effort

Caterina Pesce, Lucio Cereatti, Rita Casella, Carlo Baldari, and Laura Capranica University Institute of Motor Sciences, Rome, Italy

Running head: visual attention in older orienteers

Date of submission: November 12, 2005

Correspondence concerning this article should be addressed to: Caterina Pesce, Rome University Institute of Movement Sciences – IUSM, Piazza L. De Bosis 15, 00194 Rome, Italy Phone: +39 06 36733355, e-mail : [email protected]

Visual attentional preservation 2 Abstract This study investigated visual attention of old expert orienteers and old adults not practising activities with high attentional and psychomotor demands, and considered whether prolonged practice of orienteering may counteract the age-related deterioration of visual attentional performance both at rest and under acute exercise. In two discriminative reaction time (RT) experiments, were performed both at rest and under submaximal physical workload, visual attention was cued by means of spatial cues of different size followed, at different stimulusonset-asynchronies (SOA), by compound stimuli with local and global target features. Orienteers, as compared to non-athletes, showed a faster reaction speed and a complex pattern of attentional differences depending on the time constraints of the attentional task, the demands on endogenous attentional control, and the presence or absence of a concomitant effortful motor task. Results suggest that old expert orienteers have developed attentional skills that outweigh, at least at rest, the age-related deficits of visual attentional focusing. Key words: focus of attention, reaction time, acute and chronic exercise, aging, orienteering master

Visual attentional preservation 3 Visual attentional preservation in old expert orienteers at rest and under physical effort. Orienteering is a sport in which both endurance fitness and mental skills are needed to achieve performance. On the one side, high levels of aerobic fitness and anaerobic power are required, comparable to those of middle-distance track athletes; on the other side, attentional, decisional and problem-solving skills are essential, similar to those of sports involving information processing under time pressure (e.g., Eccles, Walsh, & Ingledew, 2002a; Kolb, Sobotka, & Werner, 1987). The main studies on expert cognition and attention in orienteering (Eccles et al., 2002a; Eccles, Walsh, & Ingledew, 2002b, 2006) are based on the analysis of verbal recall data or video, behavioral and verbal report data collected either ‘on task’ or ‘off task’. They show the complexity of the cognitive components of the orienteering performance and highlight potential contributory factors to expert cognition, such as attentional and planning strategies. In particular, the analysis of recalled performance-related information allows one to inductively identify the main task constraints of orienteering and the sequence of mental processes that must be performed to successfully cope with the cognitive task demands of orienteering. A task constraint identified as central to orienteering is the necessity to manage attention to three sources of information: the map, the environment, and travel (Eccles et al., 2002a). For an orienteer, it is critical to select relevant map information and compare it with the information from the surrounding environment. Thus, frequent switches from a narrow focus aof attention to the map (zooming in) to a wide focus of attention to the environment (zooming out) are needed. In the study of visual attentional zooming, two basic components can be differentiated: the space- and the feature- or object-based components. The space-based component refers to the selective allocation of attention to locations of different size in the visual space, whereas the object-based component refers to the

Visual attentional preservation 4 role played by attention to the selection of global and local features of visual objects (e.g., Robertson, Egly, Lamb, & Kerth, 1993). Aging seems to impair abilities involved in both the space-based and the object-based components of the attentional focusing performance: the ability to localize relevant information in the visual space in the absence of advance cues and to inhibit irrelevant information (Kahne, Hasher, Stolzfuss, & Zacks, 1994; Owsley, Burton-Danner, & Jackson, 2000), as well as the ability to ‘zoom in’ attention under narrow time constraints (Pesce, Guidetti, Baldari, Tessitore, & Capranica, 2005) and to focus on local details of visual stimuli (Roux & Ceccaldi, 2001). Because orienteering is still practiced at a competitive level at very old ages – for example, the are master classes for 90-year-olds - the age-related decline of the ability to perform visual attentional focusing might represent an important problem faced by old orienteers. However, orienteering practice couples two important factors that may counteract or offset the age-related decline of attentional performance: high physical-aerobic training and extensive experience with tasks that have high cognitive-attentional demands. A growing body of literature suggests that to induce positive effects on the cognitive and attentional functioning of older people, both cognitive training interventions and physical fitness training are appropriate (e.g., Kramer, Bherer, Colcombe, Dong, & Greenough, 2004; Roth, Goode, Clay, & Ball, 2003). Cross-sectional studies strongly suggest that age-related declines in mental efficiency can be reduced or offset by physical training (e.g., the reviews of Colcombe & Kramer, 2003; McAuley, Kramer, & Colcombe, 2004). However, results of cross-sectional studies comparing the mental performance of physically fit to that of sedentary older adults must be interpreted cautiously, as a co-variational rather than a causal relationship. In fact, individuals who elect to exercise or not exercise might differ on other socio-cultural variables that could influence mental performance (Spirduso, Francis, & MacRae,

Visual attentional preservation 5 2005). In contrast, intervention studies do not consistently demonstrate improvements in the mental performance of older individuals following exercise training. Critical variables seem to be the type and length of physical training and the particular type of cognitive functions measured (Colcombe & Kramer, 2003). For example, aerobic fitness training, but not muscle toning and weight training, produces improvements in cognitive performance. Cognitive improvements are significant even after short-term aerobic training sessions when the cognitive performances measured are selectively tied to executive control processes (e.g., Kramer et al., 1999; Moul, Goldman, & Warren, 1995). Compared to the evidence that separately concerns the effectiveness of either exercise or mental training interventions (Tomporowski, 1997), only very few studies addressed the issue of whether and to what extent the association of physical and mental training improves the cognitive functioning of older individuals (Fabre, Chamari, Mucci, Masse-Biron, & Prefault, 2002). The findings of Fabre and colleagues (2002) indicate that the indirect effects of physical fitness on cognitive-attentional processes, probably mediated by enhanced cerebral blood flow, may be significantly strengthened by coupling aerobic exercise with the direct training of cognitive processes. Nevertheless, no study specifically addresses the issue of the effects of coupling aerobic exercise training and extended practice on tasks with high visual attentional demands in older adults. In the present study, this issue is addressed by comparing visual attentional performance of older skilled orienteers and non-athletes, since orienteering naturally links together high demands on aerobic exercise and on visual attentional modulation. It can be hypothesized that the prolonged practice of orienteering either preserves visual attentional abilities from age-related deterioration, or causes the development of compensatory skills and strategies that maintain the

Visual attentional preservation 6 efficiency of those attentional processes specifically involved in the orienteering performance. Thus, the first question of this study concerns the effects of chronic aerobic exercise coupled with attention training on visual attentional performance of older orienteers. For this aim, both older expert orienteers and older non-athletes were tested on a visual attentional task previously applied to young adults (Pesce, Capranica, Tessitore, & Figura, 2003) and older individuals (Pesce et al., 2005). The task could be considered valid for two reasons. First, with respect to internal validity, in previous studies (Pesce et al., 2003), this task replicated all main effects that are reported in the literature applying similar paradigms (e.g., Robertson et al., 1993; Lamb et al., 2000); and second, with respect to external validity, task performance was sensitive both to age (Pesce et al., 2005) and to sport expertise (Pesce, Tessitore, Casella, & Capranica, in press). With special regard to the ecological component of external validity - that is to the representativeness of the task constraints - the choice of this laboratory task was made in the light of the constraints identified by Eccles and colleagues (Eccles et al., 2002a, 2006) in their research on expert cognition and attention in orienteering under real-world conditions. Since one task constraint central to their theory is attentional switching and resource sharing between three sources of information (i.e., the map, the environment, and the travel), it is likely that a laboratory task that involves zooming attention between different spatial scales and visual object levels is sufficiently representative of the real task of interest. In the chosen attentional focusing task, the space-based and object-based components of visual attentional focusing were investigated in combination to verify whether either one or both components are preserved in old orienteers. The space-based aspect of the attentional focusing was operationalized by means of the effects of cueing attention with cues of different size, whereas the object-based aspect was operationalized by means of the effects of presenting compound stimuli with global and local target features. Also, two critical variables that strongly influence the

Visual attentional preservation 7 visual attentional performance of older adults are the time constraints of the task and the demands on endogenous (i.e., top-down) control of attention, which reflect the difficulty of older adults to intentionally modulate visual attention under high time pressure (Pesce et al., 2005). Thus, these variables were also considered. However to limit the complexity of the factorial design, the demands on endogenous attentional control were manipulated separately, thus generating two different experiments. Besides the effects of chronic attentional and physical exercise training on visual attentional performance of older skilled orienteers, the present study also investigates the effects of acute physical exercise on the attentional performance. The literature on younger adults indicates that acute exercise may have beneficial or detrimental effects on concomitant cognitive and attentional performances depending on exercise type and intensity, as well as on the type and difficulty of the cognitive or attentional task (Brisswalter, Collardeau, & René, 2002; McMorris & Graydon, 2000; Tomporowski, 2003). Submaximal, constant workloads seem to have beneficial effects on the visual attentional performance of physically active young adults (Pesce et al., 2002, 2003). Particularly, physical effort might enhance attentional performance by increasing the amount of resources that may be allocated to task specific demands (e.g., Delignières, Brisswalter, & Legros, 1994). With specific regard to orienteering, few studies investigated the effects of acute physical effort on visual performance. Arcelin, Fleurance, and Brisswalter (1995) showed a facilitation of visual information processing speed without any change in the speed-accuracy trade-off setpoint under workloads up to 90% VO2max. In contrast, Hancock and McNaughton (1986) showed an impairment of visual performance accuracy under workloads above the anaerobic threshold. However, the different methodological approaches and low sample sizes do not allow thorough comparisons and definitive conclusions.

Visual attentional preservation 8 Because no study has addressed the effects of physical effort on the visual attentional performance of sedentary or physically and cognitively trained older adults, the hypothesis of the present study related to this second question is that older orienteers, who are used to modulating visual attention under submaximal physical effort, may maintain beneficial effects of submaximal efforts on attentional performances similar to those observed in younger adults. Recent research on visual attention in both young and adult orienteers (Eccles et al., 2006) showed that more experienced athletes allocate attention to the map while moving more often than those who are less experienced, thus, being better able to strategically control attention while performing an effortful physical task. In the case of older non-orienteers, who are not accustomed to coupling physical effort and attentional psychomotor performance, a negative dual task effect might emerge (Brisswalter et al., 2002). Acute physical exercise was operationalized by means of the effects of a submaximal physical load on cylo-ergometer (60% heart rate reserve [HRR]) on concomitant attentional task performance. Experiment 1 Method Participants. Twenty-five physically active participants (12 male and female old orienteers and 13 male and female controls), aged 60-75 years (orienteers: M = 66.2 ± 4.7 years, controls: M = 66.3 ± 4.6 years), provided informed consent to take part in the experiment. The criterion for inclusion in the subgroup of athletes was that they were club-standard orienteers participating in National and/or International competitions with at least 10 years’ experience. Prior to the experiment, participants answered the AAHPERD exercise/medical history questionnaire to ascertain his or her activity level, educational background, smoking and dietary habits, medication use and history of physical activity. Individuals with any of the following conditions were excluded from the study: evidence or known

Visual attentional preservation 9 history of neuromuscular disorders, cognitive impairment, or use of medications that would affect the test performance. All participants were middle-class and lived a fully independent and noninstitutionalized lifestyle. All of them were right-handed, had a normal or corrected-to-normal near and far visual acuity, were unaware of the purpose of the experiment and didn’t have any prior experience with the test setting. Apparatus and stimuli. The participants were seated in a dimly lit room at a distance of 60 cm from a PC-driven video screen, with their eyes at a height corresponding to the horizontal axis crossing the centre of the screen. Four types of visual displays were presented on the screen (Figure 1): an instruction, a central fixation point, a spatial cue of variable size, and a compound stimulus. The fixation point was a tilted “T” of 0.4° x 0.4° and the spatial cue was an empty box of 1° x 1° or 5° x 5°. The compound stimulus was a large letter (4.6° x 4.6°) made of 13-17 small letters (0.6° x 0.6°) spaced 0.4° in a 5 x 5 matrix (Figure 2). The large letter and its small elements represented the global and local level of the compound stimulus, respectively. The large letter could be an A, E, F or H; the small letters were the remaining letters (e.g., a global "H" letter composed of local "A", "E" and “F” letters). The fixation point, the large box and the following compound stimulus were centered on the screen; the small box could randomly appear at one of the 13-17 locations of the local elements composing the compound stimulus that followed. Insert figures 1-2 about here The attentional task. At the beginning of each block of trials, the display informed participants about the task and the target stimulus to be discriminated (Figure 1); after a self-paced interval, participants cleared the screen for the first trial to begin. Each trial consisted of the following sequence of events. First, the fixation point appeared and then, after a 500 ms interval, a large or a small cue was shown for either 70 or 420 ms. Thereafter, a blank field of 80 ms duration separated

Visual attentional preservation 10 the offset of the cue from the onset of the upcoming compound stimulus to avoid forward masking. Thus, the time interval between onset of the cue and onset of the compound stimulus (cue-target stimulus-onset-asynchronies [SOA]) was of either 150 or 500 ms. In five sixths of the trials, the compound stimulus contained the predefined target letter (e.g., "H") either at the global or at the local level (Figure 2). In the remaining trials, the compound stimulus did not contain the target letter (catch trials). The frequency of global or local target occurrence within each block was set at 50% to balance the priming effects between consecutive global- or local-target trials (Robertson, 1996). In 80% of the trials, there was an isomorphic relation between the size of the cue and the size of the upcoming target it signaled: a large cue was followed by a global target (Figure 2, panel a) and a small cue by a local target (Figure 2, panel c). In the remaining 20% of trials, cue and target size were mismatched (Figure 2, panels b and d). Before the experiment, the participants were instructed to focus their attention on the area of the visual field delimited by the spatial cue, without shifting their gaze, in order to react as soon as possible to a predefined target letter that would probably match cue size. The reaction consisted in pressing a reaction-time (RT) key aligned to the body midline with the right index finger. Further instructions were aimed at forcing participants, in the case of cue-target mismatching, to directly switch from the global to the local level (attentional zooming in) or from the local to the global level (zooming out), avoiding visual search strategies. It was explained that when a large cue was not followed by a global target letter, the target was the local letter at the center of the screen (Figure 2, panel b); when a small cue was not followed by a local target letter at the cued location, the target was the global letter (Figure 2, panel d). The participants had to refrain from responding when the target letter was not present (catch trials). Responses to catch trials or responses with RTs shorter than 200 ms or longer than 2,500 ms were considered errors (anticipations and

Visual attentional preservation 11 delayed responses, respectively) and were discarded. The response caused the offset of the compound stimulus for the next trial to begin after an intertrial interval of 1,000 ms. The attentional task was divided into two blocks of 76 trials each, one with shorter SOA and one with longer SOA, respectively. They order was counterbalanced across participants. Each block, preceded by one block of 40 practice trials, included 4 warm-up trials, 12 catch trials and 60 full experimental trials. Cue sizes and target levels were balanced and randomised within blocks. According to cue probability (i.e., 80% valid cues and 20% invalid cues), the 60 full experimental trials were divided as follows: 48 trials with cue-target matching and 12 trials with cue-target mismatching. To reduce potential threats to internal validity deriving from the use of four different letters, the following counterbalancing and randomisation were applied. Four blocks—one for each possible target letter—were combined pairwise, and participants were randomly assigned to one of the resulting six possible block combinations: A-E, A-F, A-H, E-F, E-H, or F-H. Furthermore, for each target letter, all possible combinations with the remaining nontarget letters at the global and local level of the compound stimuli were balanced and randomized within blocks (e.g., a small target letter “A” could be equally included in a large E composed of small Fs and Hs, or in a large F composed of small Es and Hs, or in a large H composed of Es and Fs). Physical exercise. In separate sessions, participants performed the attentional task twice, at rest and under constant 60% maximal oxygen uptake (VO2max) cycling at 50-60 rev/min, respectively. Individual workload intensities were determined as the workload corresponding to a target heart rate of 60% heart rate reserve (HRR) (Karvonen, Kentala, & Mustala, 1957). It has be pointed out that percentage of HRR is not equivalent to percentage of VO2max, but represents an underestimate of percentage of VO2max especially for low-fit individuals at low exercise intensities

Visual attentional preservation 12 (e.g., Swain, 2000). Thus, the chosen target heart rate of 60% HRR might correspond to different exercise intensities depending on differences in fitness level of skilled older orienteers and nonathletes. However, avoiding interindividual differences in exercise intensity is less relevant to the aim of the present study than ensuring a similar relative increase in effort intensity above rest for participants of different fitness levels. For this purpose, percentage of HRR and the linearly correlated percentage of VO2 reserve (%VO2R) are more appropriate parameters than percentage of VO2max. In fact, percentage of VO2max at rest varies inversely with fitness level, and a target exercise intensity of given percentage of VO2max represents a smaller relative increase above rest for a less-fit participant than for a more-fit one, while resting VO2 and HR are 0% of VO2R and HRR for less-fit as well as for more-fit participants, and an exercise intensity of given percentage of VO2R or percentage of HRR represents a fixed relative increase above rest independently of fitness level (e.g., Swain, 2000). Each of the two blocks of experimental trials lasted 3-4 minutes, and the experimental session at rest (including instruction and training block) lasted about 12 minutes. The overall duration of the session under physical load lasted longer because participants warmed up to their individual heart rates according to an incremental protocol that consisted of 2 minute steps with a load increment of 50 W starting with a 50-W load. Then, the attentional task was started, and continuous exercise was done by cycling at 50-60/rev.min with heart rate monitoring throughout. During the attentional test, adjustments of the individual workload were done if the actual heart rate deviated from target heart rate more than ± 5 beat/min. To minimize learning effects, the sessions at rest and under physical workload were separated by 1 to 3 months, and to exclude the effects of physical fatigue on attentional performance, participants were asked to refrain from intensive training on the day before the experimental

Visual attentional preservation 13 session. Furthermore, the sequence of the sessions at rest and during exercise was counterbalanced across participants in order to uniformly distribute eventual learning/fatigue effects across resting and working conditions. Preliminary analyses Both error data (anticipations and delayed responses) and RTs were collected. Percentages of delayed responses (RTs longer than 2,500 ms) and of anticipations (RTs shorter than 200 ms and reactions on catch trials), and median RTs for correct responses were separately calculated for orienteers and nonathletes, both at rest and under physical load (Table 1). Median RTs were calculated instead of the more common mean RTs because the frequency of outliers that disproportionally contribute on mean RTs may be different in older orienteers vs. nonathletes, with higher rates probable in older nonathletes. Comparing RTs and error rates of orienteers and nonathletes (Table 1) indicates that differences in reaction speed were not due to a different speedaccuracy trade-off setpoint. Thus, only RTs were analyzed further. Insert Table 1 about here Results and Discussion Median RTs were submitted to a 2 x 2 x 2 x 2 x 2 ANOVA with repeated measures on four of the five factors. The between-participants factor was group (orienteers vs. nonathletes). The withinparticipants factors were physical load (rest vs. sub-maximal workload), cue-target SOA (150 vs. 500 ms), cue size (large vs. small), and target level (global vs. local). Post-hoc analyses were performed by means of planned pairwise comparisons (t tests). Since our hypotheses were based on interaction effects of higher order, to eliminate the problem of an inflated Type 1 error as a result of multiple comparisons, alpha levels was adjusted by means of the Bonferroni technique.

Visual attentional preservation 14 A main effect for group emerged, F(1, 23) = 7.04, p < .016, indicating that older orienteers had faster RTs than older nonathletes (679 ± 39 vs. 810 ± 30 ms). Furthermore, there was a main effect for SOA, F(1, 23) = 6.58, p = .019, and the following significant interactions: SOA x Target Level, F(1, 23) = 8.71, p = .008, Cue Size x Target Level, F(1, 23) = 11.04, p = .004, and SOA x Cue Size x Target Level, F(1, 23) = 13.15, p = .002. Also, the attentional factors (SOA, cue size, and target level) interacted with group and physical load: Group x Physical Load x SOA x Cue Size x Target Level, F(1, 23) = 4.87, p = .040. Because the previous two- and three-way interactions are included in the five-way interaction, only the last one will be described. This interaction is relevant to the focus of the present study because it indicates that older orienteers focus visual attention differently than older nonathletes, and that their way to focus attention also differs as a function of whether they perform the task at rest or under a concomitant physical effort. In the post-hoc analysis, simple effects of the Cue Size x Target Level interaction, which are indicators of the attentional zooming in/out, were separately computed for orienteers and non-athletes, both at rest and under physical load, with shorter and longer SOA (for eight comparisons, adjusted p < .006). Means and standard deviations (SD) of the corresponding median RTs are reported on Table 2. The attentional performance of older orienteers and older nonathletes significantly differed only under narrow time constraints (i.e., shorter SOA). At longer SOA, both at rest and under physical load, older orienteers as well older nonathletes showed a pattern of attentional effects similar to that observed in younger adults (Pesce et al., 2003, 2005): They reacted faster to global targets than to local targets when the cue was large but, conversely, reacted faster to local targets than to global targets when the cue was small. This result may be graphically represented in terms of global-local RT differences (Figures 3 and 4). Cueing attention to focus at the global level (i.e., large cue)

Visual attentional preservation 15 worsened the discrimination of local targets (RT cost in Figure 3, which is significant at rest for both orienteers and nonathletes: t(11) = -4.28, p = .001, and t(11) = -5.68, p = .0001, respectively). Conversely, cueing attention to focus at the local level (i.e., small cue), enhanced the discrimination of local targets (RT benefit in Figure 4, which is significant under workload for both orienteers and nonathletes: t(11) = 4.10, p = .002, and t(11) = 5.52, p = .0002, respectively, and at rest only for nonathletes: t(11) = 9.94, p