Visual search is the primary means by which people acquire information from

Transcription

1 Search Performance of Healthy Adults on Cancellation Tests Mary Warren, Jennifer M. Moore, Laura K. Vogtle KEY WORDS aging attention brain injuries rehabilitation visual perception Cancellation tests are used extensively to identify visual search deficiencies in people with neurologic conditions, but little is known about how healthy adults perform on these tests. This study described the performance and types of search strategies used by healthy adults to complete cancellation tests from the Brain Injury Visual Assessment Battery for Adults. Study participants predominantly used structured left-to-right and top-to-bottom linear search patterns regardless of the configuration of the visual array. Other search qualities included (1) use of the same pattern throughout the search of an array, (2) symmetrical search of the array, and (3) rechecking performance on complex arrays. Older adult participants took more time to complete the tests than younger adults did. Knowing the strategies healthy adults use on cancellation tests enables therapists to identify visual search deficiencies in clients and design interventions to reestablish normal search strategies in daily occupations. Warren, M., Moore, J. M., & Vogtle, L. K. (2008). Search performance of healthy adults on cancellation tests. American Journal of Occupational Therapy, 62, Mary Warren, OTR/L, SCLV, FAOTA, is Associate Professor, Department of Occupational Therapy, University of Alabama at Birmingham, rd Avenue South, Birmingham, AL ; Jennifer M. Moore, MS, OTR/L, is Occupational Therapist, Medical West: An Affiliate of the University of Alabama at Birmingham Health System, Bessemer, Alabama. Laura K. Vogtle, PhD, OTR/L, is Professor, Department of Occupational Therapy, University of Alabama at Birmingham, Birmingham, Alabama. Visual search is the primary means by which people acquire information from the environment, and it is executed in a systematic and organized fashion to locate objects in the least amount of time (Mort & Kennard, 2003). Acquired brain injury and neurologic disease often alter the quality and comprehensiveness of visual search (Ebersback, Trottenberg, Schelosky, Schraq, & Poewe, 1996; Kerkhoff, 2001; Mendez, Cherrier, & Cymerman, 1997; Mort & Kennard, 2003; Zihl, 2000). Hemianopsia and visual spatial neglect, two conditions common in adults with acquired brain injury, have been shown to significantly disrupt search patterns. Hemianopsia can impair ability to search the blind side of the visual field, causing an abbreviated search pattern characterized by omissions (Zihl, 2000). Visual spatial neglect disrupts the mechanisms that direct visual attention, resulting in unsystematic and abbreviated search of the contralesional side (Mort & Kennard, 2003). Alzheimer s and Parkinson s disease have also been shown to disrupt visual search (Ebersback et al., 1996; Mendez et al., 1997). The person environment occupation model (Law, Cooper, Strong, Stewart, Rigby et al., 1996) forms the theoretical foundation for the application of visual search to practice. The essence of that theory is that occupational performance is based on dynamic interactions between the person, task, and environment. Change in any one of the three requires adaptation and modification to ensure satisfactory occupational performance. Occupational therapists can use this model to assess alterations in the interaction between the three entities and design interventions to provide new strategies or adjustments to improve occupational performance. Impaired visual search affects completion of daily occupations through its influence on reading and mobility (Zihl, 2000). People with hemianopsia or hemiinattention omit letters and words on the impaired side, which reduces reading rate, accuracy, and comprehension (Kerkhoff, 2001; Zihl, 1995). The subsequent reading deficiency affects ability to complete daily occupations that incorporate reading, such as financial management, shopping, and cooking. These conditions also affect 588 September/October 2008, Volume 62, Number 5

2 mobility when the person does not search the environment completely to locate objects and obstacles, which contributes to deficiencies in driving and safe navigation of environments (Kerkhoff, 2001; Pambakian et al., 2000). Because of the potential disruption to completion of daily occupations, occupational therapists routinely assess visual search capability in clients with brain injuries and neurologic disease. Paper-and-pencil cancellation tests are commonly used to evaluate visual search performance (Uttl & Pilkenton-Taylor, 2001). These tests consist of strings of letters or symbols with targets embedded throughout the array (Figure 1). The client is instructed to scan the array and cross out as many targets as possible. Test performance is scored for target errors and is usually timed. Several evaluations used by occupational therapists to assess visual perceptual processing incorporate cancellation tests, including the Behavioral Inattention Test (Wilson, Cockburn, & Halligan, 1987), the Motor-Free Visual Perception Test (Colarusso & Hammill, 2003), and the Brain Injury Visual Assessment Battery for Adults (bivaba; Warren, 1998). Figure 1. bivaba Visual Search Subtests. From Warren, M. (1998). Brain Injury Visual Assessment Battery for Adults Test Manual, pp Birmingham, AL: visabilities Rehab Services. Copyright 1998 by visabilities Rehab Services. Reprinted with permission. The American Journal of Occupational Therapy 589

3 All of the cancellation tests in these evaluations record completion time and error rate, but the bivaba also identifies the type of search strategy the client uses to locate targets and documents deviations associated with hemianopsia and neglect. According to Warren (1998), the bivaba cancellation tests include analysis of the search pattern because it provides more useful information for intervention planning than tests that record only time and errors. Knowing specifically how the search pattern has changed enables the therapist to design specific interventions to reestablish a more effective search strategy in daily occupations. Use of the task analysis approach proposed by Warren (1998) is predicated on understanding how healthy adults search for targets on cancellation tasks and their typical test performance. Although cancellation tests are used extensively to identify visual search deficiencies in people with neurologic conditions, few studies have investigated how healthy adults perform on these tests, and completed studies focus exclusively on speed and accuracy (Byrd, Touradji, Tang, & Manly, 2004; Geldmacher & Riedel, 1999; Uttl & Pilkenton- Taylor, 2001). The purpose of this pilot study was to investigate the predominant search patterns used by adult men and women to complete cancellation tests. The study expanded on data previously collected by Warren (1998) on adult performance on the seven visual search subtests of the bivaba (Figure 1) and followed the protocol established for these tests. The study hypotheses were as follows: (1) Adults will predominantly use a left-to-right reading search strategy, (2) there will be no difference between men and women in subtest performance, (3) older adults will require more time than younger adults to complete the subtests, and (4) participants will check their work after completing the subtests. Method Sample A convenience sample of healthy adults was recruited from the community. The sample included men and women for each 10-year increment between 20 and 90 years of age (Table 1). Exclusion criteria included history or presence of the following: central visual field impairment, acuity less than 20/100, macular degeneration, diabetic retinopathy, brain injury, learning disability, attention deficit disorder, or physical impairment preventing accurate target cancellation. The university s institutional review board approved the proposal, and all participants signed consent forms. Instrument The bivaba consists of standardized assessments that measure acuity, contrast sensitivity, oculomotor function, visual Table 1. Sample Characteristics field, and visual search (Warren, 1998). The seven visual search subtests use a cancellation format and consist of 8½- 14-in. sheets of paper with letters, circles, or geometric shapes in either linear or random arrays (Figure 1). For all subtests except Subtests 5 and 6, the participant was instructed to search the visual array and draw a line through the designated target. On Subtests 5 and 6, the participant was instructed to number the open circles sequentially from 1 to 20 (Subtest 5) or 1 to 40 (Subtest 6). Procedure Men Women White Other Ethnicity years years years years years years years Total Participants completed the subtests in a single session seated at a table in a quiet, well-lit room. The investigator presented the subtests one at a time in the sequence given in Table 2. The subtest was placed on the table at the participant s midline. Before beginning each subtest, the investigator showed the participant the target(s) located on the subtest at the top of the page. The investigator assured the participant that there was no correct way to search for targets and that although the test would be timed, he or she could take as much time as needed to find all of the targets. The participant used a red marker to cross out the targets. Timing with a stopwatch began when the participant picked up the marker and stopped when the participant laid it down on the table. During the test, the investigator observed and recorded the predominant search pattern the participant used to locate targets. Completion time and the number of errors of omission and incorrect target were recorded. From earlier pilot testing, it was expected that the participant would use one of four commonly used search patterns (Figure 2) to complete the subtests: (1) a horizontal, left-toright, top-to-bottom, linear (reading) pattern; (2) an alternating horizontal left-to-right and right-to-left, top-to-bottom linear pattern, designated as a horizontal rectilinear pattern; (3) an alternating vertical left-to-right and right-toleft top-to-bottom linear pattern designated as a vertical rectilinear pattern; and (4) a vertical left-to-right, top-to-bottom linear pattern. The recording form listed the four patterns with space to record deviations from the patterns. The 590 September/October 2008, Volume 62, Number 5

4 Table 2. Frequency of Visual Search Strategies and Checking Work on the Brain Injury Visual Assessment Battery for Adults Subtests Switches (%) Search Strategy Option Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Horizontal left to right Horizontal rectilinear Vertical left to right Vertical rectilinear Vertical right to left Diagonal Circular Bottom to top Switches strategies Checked work investigator also recorded whether the participant rescanned the test to check the accuracy of the performance. Results The study sample consisted of 81 participants: 73 White, 1 Hispanic, 1 Asian, and 6 African American male and female participants between 20 and 86 years of age (see Table 1). All participants had completed high school, and several had postsecondary degrees. The investigators analyzed the data to assess performance differences and relationships between age groups, gender, and ethnicity using the Statistical Package for the Social Sciences (SPSS Version 11, 2002; SPSS, Inc., Chicago). The data analysis included descriptive statistics, two-factor analysis of variance (ANOVA), and post hoc tests to test for differences between age, gender, and total test time; and chi-square analysis to test for independence in search pattern, age, and gender. Age data were recoded into three groups: years, years, and years for the analyses. Table 2 shows the percentage of participants using each visual search strategy and the percentage who checked their work on the seven subtests across all age groups. Chi-square analysis was completed to identify the most prevalent pattern for each subtest (Table 3). The horizontal, left-to-right, top-to-bottom reading pattern was found to be the predominant search strategy used by participants on all subtests, followed by the horizontal rectilinear (alternating) pattern. In addition to the four expected patterns (Figure 2), four other patterns were observed: vertical right to left, diagonal, circular, and bottom to top (Table 2). The percentage of participants using these alternate patterns was very small (1.2% 3.8%). All participants used structured linear search patterns on the subtests with characters arraigned in rows and columns (Subtests 1 4). Most participants applied the same search strategy throughout a subtest; however, a small number switched strategies during the test. The incidence of switching strategies ranged from 2.5% on Subtest 3 to 17.5% on Subtest 7. Independence of gender and search pattern and of age and search pattern was tested Table 3. Use of Search Pattern and Frequency of Checking Work on the Brain Injury Visual Assessment Battery for Adults Subtests Figure 2. Commonly used search patterns on the bivaba. From Warren, M. (1998). Brain Injury Visual Assessment Battery for Adults Test Manual, pp Birmingham, AL: visabilities Rehab Services. Copyright 1998 by visabilities Rehab Services. Reprinted with permission. c 2 df p < c 2 df p < Test * Test * Test Test Test Test Test Note. N = 8.1 *Not significant. The American Journal of Occupational Therapy 591

5 using a two-factor chi-square analysis. No difference was found between men and women or between the three age groups in the predominant search pattern used for each subtest. Table 4 shows the mean time required to complete each subtest. Standard deviations indicated a wide variability in completion times. Participants completed the 20-item open-circle numbering subtest (5) in the least amount of time (average 23.5 s). The maximum amount of time required to complete a subtest was 215 s (3.5 min) and occurred on Subtest 2, which requires cancellation of singleletter targets among crowded alphabet string arrays (Table 4). A two-factor fixed ANOVA was used to test the differences between age, gender, and total time required to complete the subtests. A main effect for age was found (F[2] = , p <.000). Post hoc analysis using a Student Newman Keuls test showed that participants 60 years and older took a longer time to complete the subtests than those between 20 and 59 years. The frequency of participants checking their work for accuracy ranged from 1.2% on Subtest 5 to 39.5% on Subtest 1 (Table 2). Chi-square analysis revealed that participants checked their work less frequently than expected on all subtests except Subtests 1 and 2 (Table 3). The number of participants checking work declined with each subtest into single digits until the last subtest (7), when the percentage increased again into double digits. The increase may have occurred because Subtest 7 has the most complex and least structured visual array of the subtests. A two-factor chisquare analysis testing independence of age and checking work (Table 3) showed no difference between age groups except on Subtest 7, where older adults were more likely to check their work than younger adults (χ 2 [2] = 6.466, p <.039). A two-factor chi-square analysis testing independence between gender and frequency of checking work found no significant difference between men and women. Table 4 summarizes the types of errors committed on the subtests. Omission (missed target) was the most common error on all subtests. Participants committed the most errors on Subtest 2, which involved locating letters in a dense array of alphabet strings, and the least errors on Subtests 5 and 6 where participants sequentially numbered circles. Mean scores and standard deviations for the subtests showed that participants demonstrated a high degree of accuracy with minimum variability. A two-factor fixed ANOVA testing the differences between age, gender, and total score on the subtests was not statistically significant, indicating that age and gender did not influence the error rate of the participants. Discussion The findings of this study support the consensus among researchers that visual search is completed in an organized, systematic, and timely manner in healthy adults without brain injury (Mort & Kennard, 2003). The study showed that participants exhibited seven characteristics in searching cancellation tests: (1) use of structured patterns, (2) strong tendency to scan left to right and top to bottom, (3) consistent employment of one pattern throughout search of the visual array, (4) few errors, (5) symmetrical scanning of the array, (6) completion times of less than 4 min per test, and (7) rechecking of work on complex arrays. With few exceptions, participants executed structured linear patterns that were predominantly left-to-right and top-to-bottom on all subtests regardless of the structure of the visual array. The strong preference for this search pattern may be because it is used to read English and thus constantly reinforced. Although participants tended to use the familiar left-toright search pattern, the study results suggest a willingness to modify or use certain search patterns to gain advantage in test performance. For example, the second most frequently used pattern, the horizontal back-and-forth configuration, eliminates the long leftward saccade to locate the next line and therefore might have been chosen to increase performance speed by participants who were conscious of being timed. Once a strategy was selected, participants tended to use it throughout the subtest. On Subtest 7, however, the Table 4. Number of Targets, Mean Score, Mean Time, and Percentage of Different Errors Committed on Brain Injury Visual Assessment Battery for Adults Subtests Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Targets Mean score SD Mean time, s SD Omission (%) Incorrect target (%) Omission and incorrect target (%) September/October 2008, Volume 62, Number 5

6 random complex circles test, 17.5% of the participants changed search patterns during the subtest. More participants may have changed search patterns on this test because it has the least structured and most complex visual array and they switched in an effort to use a more efficient search strategy. No age or gender difference was noted among participants who deviated from predominant patterns, and the accuracy rate was high regardless of whether the participant switched search patterns. The finding that participants older than age 60 required significantly more time to complete the subtests is consistent with previous investigations (Hommel, Li, & Li, 2004; Uttl & Pilkenton-Taylor, 2001). According to Hommel et al. (2004), cautiousness and age-related changes in cognition contribute to the increased time required by older adults to complete cancellation tests. No age difference was observed in error rate, however, and all participants demonstrated a high rate of accuracy on the subtests (Table 4). The largest number of errors occurred on the first subtests (Table 4) and then steadily declined, suggesting that unfamiliarity with the task may have been cause. The increase in error rate on the last subtest (7) may have occurred because of its visual complexity. Errors were evenly distributed throughout each subtest, suggesting that participants searched the arrays symmetrically. The finding that few participants rechecked their work (Table 2) was unexpected, because intuitively it seems that people would routinely check work to ensure accuracy. Nearly 40% of the participants checked their work on the first subtest, checking less frequently on each subsequent subtest until Subtest 7, when the frequency again increased. This finding suggests that as people become more comfortable with a search task, they feel less need to check their work unless they encounter a more complex or unusual visual array. The finding that more older adults checked their work on Subtest 7 than younger adults suggests that older adults may be less sure of their performance on complex arrays. Application to Occupational Therapy Practice When visual search is disrupted by brain injury or disease, the person does not take in visual information in a systematic fashion that can be used for tasks that make up occupational performance. Applying therapeutic interventions that ensure that visual information is taken in through the use of structured, efficient search patterns should improve occupational performance. The findings of this study suggest that healthy adults can be expected to predominantly use a left-to-right, top-to-bottom reading search strategy to complete cancellation tests. The selected search pattern will be consistently used throughout testing, but the person may switch patterns if it is perceived that an advantage can be gained in test performance. The search pattern will be symmetrical and result in a high rate of accuracy and good speed, although older adults will likely require increased time. The more complex the array being searched the more likely the person will recheck work to accuracy. Persons with hemianopsia, hemi-inattention, or neglect search much differently on cancellation tasks. People with hemianopsia often abbreviate searches to the blind side, causing them to omit targets (Zihl, 2000). However, they execute an organized left-to-right pattern during visual search and accurately cancel targets on the side of the visual array that they are able to see (Zihl, 2000). In contrast, people with hemi-inattention and neglect from right hemisphere lesions demonstrate unorganized and random search strategies across the visual array. They often search only for targets on the right side of the array, an approach that results in significant target omissions to the left; they also often misidentify targets (Kerkhoff, 2001; Weintraub & Mesulam, 1988). Knowing the expected performance on a cancellation task enables therapists to identify the strengths and limitations of the client s search performance. For example, observing that a client with a left hemianopsia uses the expected left-to-right search pattern to locate targets but omits targets on the left side of the test suggests that the client s ability to initiate a normal search pattern has not been disrupted but that the client does not see all of the visual array on the left side. The therapist would direct intervention toward enabling the client to scan further to the left side when reading or searching for items in the environment. If instead the therapist observes that the client initiates visual search from the bottom of the right side, scans from the right to the left, and abbreviates search on the left side, intervention should be directed toward establishing a structured left-to-right search strategy combined with scanning further to the left side. In addition, knowing that older adults take longer to complete cancellation tests but are able to complete a test in a certain amount of time enables the therapist to determine whether the longer time required by an older client is indicative of age or impairment. In all cases, the therapist, by analyzing the client s search performance and comparing it to the expected performance, is able to discern the most appropriate intervention strategy. Study Limitations and Future Research Limitations of the study included use of a small convenience sample with limited racial and cultural diversity. At least one study has shown that ethnicity may affect speed and accuracy on cancellation test performance (Byrd et al., 2004). It also would have been beneficial to interview participants as to why they chose certain search patterns or modified performance The American Journal of Occupational Therapy 593

7 to better understand their behavior on the tests. Additional research is needed to determine whether race, culture, or predominant language influence performance in healthy adults and to explicitly delineate the patterns used by persons with brain injury and neurologic conditions. s References Byrd, D. A., Touradji, P., Tang, M., & Manly, J. J. (2004). Cancellation test performance in African American, Hispanic, and White elderly. Journal of the International Neuropsychological Society, 10, Colarusso, R., & Hammill, D. (2003). Motor-Free Visual Perceptual Test 3. Novato, CA: Academic Therapy. Ebersback, G., Trottenberg, T., Schelosky, L., Schraq, A., & Poewe, W. (1996). Directional bias of initial visual exploration: A symptom of neglect in Parkinson s disease. Brain, 111, Geldmacher, D. S., & Riedel, T. M. (1999). Age effects on random-array letter cancellation tests. Neuropsychiatry, Neuropsychology, and Behavioral Neurology, 12, Hommel, B., Li, K. Z., & Li, S. (2004). Visual search across the life span. Developmental Psychology, 40, Kerkhoff, G. (2001). Spatial hemineglect in humans. Progress in Neurobiology, 63, Law, M., Cooper, B., Strong, S., Stewart, D., Rigby, P., & Letts, L. (1996). The person environment occupation model: A transactive approach to occupational performance. Canadian Journal of Occupational Therapy, 63, Mendez, M. F., Cherrier, M. M., & Cymerman, J. S. (1997). Hemispatial neglect on visual search tasks in Alzheimer s disease. Neuropsychiatry, Neuropsychology, and Behavioral Neurology, 10, Mort, D. J., & Kennard, C. (2003). Visual search and its disorders. Current Opinion in Neurology, 16, Pambakian, A. L. M., Wooding, D. S., Patel, N., Morland, A. B., Kennard, C., & Mannan, S. K. (2000). Scanning the visual world: A study of patients with homonymous hemianopia. Journal of Neurology, Neurosurgery, and Psychiatry, 69, Uttl, B., & Pilkenton-Taylor, C. (2001). Letter cancellation performance across the adult life span. Clinical Neuropsychologist, 15, Warren, M. (1998). Brain Injury Visual Assessment Battery for Adults test manual. Birmingham, AL: visabilities Rehab Services. Weintraub, S., & Mesulam, M. M. (1988). Visual hemispatial inattention: Stimulus parameters and exploratory strategies. Journal of Neurology, Neurosurgery, and Psychiatry, 51, Wilson, B. A., Cockburn, J., & Halligan, P. W. (1987). Behavioral Inattention Test. Fareham, England: Thames Valley Test Co., & Los Angeles: Western Psychological. Zihl, J. (1995). Eye movement patterns in hemianopic dyslexia, Brain, 118, Zihl, J. (2000). Rehabilitation of visual disorders after brain injury. Hove, England: Psychology Press. REVISED EDITION OF LOW VISION SPCC! Low Vision: Occupational Therapy Evaluation and Intervention With Older Adults Self-Paced Clinical Course (SPCC) Edited by Mary Warren, MS, OTR/L, SCLV, FAOTA Earn 2 AOTA CEUs (20 NBCOT PDUs/20 contact hours) Order #3025-P Member Price: $370 Non-Member Price: $470 When the first edition of the AOTA Low Vision SPCC was published in 2000, occupational therapy practice in low vision rehabilitation services was relatively new. What a difference 8 years makes! Significant transformations in the role of occupational therapy in this practice area and the increasing need for services require that occupational therapy practitioners in low vision rehabilitation maintain professional competency. To meet this demand, AOTA established specialty certification in low vision rehabilitation (SCLV) in This revised edition of the Low Vision SPCC supports the SCLV certification process and includes revisions, updates, new information on evaluation, and lessons related to psychosocial issues and low vision. SPCC Highlights Outlines low vision rehabilitation and the role of occupational therapy Presents detailed lessons in 10 subject areas that support specialty certification eligibility Examines eye conditions that cause low vision in adults Provides instruction on basic optics and optical devices Communicates the importance of evaluation, intervention, and follow-up care Call AOTA Shop online at store.aota.org (Self-Paced Clinical Course) SP September/October 2008, Volume 62, Number 5