Continuous performance tests in schizophrenic patients: Stimulus and medication effects on performance

PsychiatryResearch, 31~47-56 Elsevier

Continuous Stimulus Elizabeth Philip

47

Performance Tests in Schizophrenic Patients: and Medication Effects on Performance

A. Earle-Boyer, D. Harvey

Mark

R. Serper,

Michael

Davidson,

and

Received Ju!v 23, 1990; revised version received December 27, 1990; accepted Februar_v 3. 1991. Abstract. Medicated (n = 17) and unmedicated (n = 17) schizophrenic patients were compared to a normal control group (n = 19) on their performance on auditory and visual versions of the Continuous Performance Test (CPT). Within each stimulus modality, performance was examined on lexical and nonlexical target stimuli. Neuromotor competence was assessed on the basis of motor speed and proficiency. Normal subjects made fewer errors of all types than schizophrenic patients. Unmedicated patients made significantly more errors on nonlexical stimuli than medicated patients, with medication status found not to be associated with stimulus modality effects. Motor proficiency was associated with CPT performance in the medicated patients, but not the unmedicated ones, although this difference in correlations did not account for the group differences between these patients. The authors discuss the implications of these data for the type of cognitive and attentional functions that are affected by medication in schizophrenia.

Key Words. Vigilance, neuroleptics,

motor performance,

negative

symptoms.

Attentional deficits have been suggested to be common in schizophrenic patients since the first conception of the disorder (Bleuler, 191 l/ 1950; Kraepelin, 1919). Schizophrenic performance deficits on a wide variety of laboratory measures of attentional functioning have been identified. These include problems in the following areas: (1) the ability to attend to target information in the presence of distracting stimuli (selective attention; Oltmanns, 1978; Rund, 1983; Harvey et al., 1990); (2) identification of visual target information in the context of a display containing irrelevant information (Span of Apprehension; Neale et al., 1969; Asarnow and MacCrimmon, 1978); and (3) the ability to process target information followed by a noninformative masking stimulus (Backward Masking; Saccuzzo et al., 1974).

Elizabeth A. Earle-Boyer, Ph.D., was Research Associate, Mt. Sinai School of Medicine, New York, NY. Mark R. &per, M.A., is Research Fellow, Mt. Sinai School of Medicine, and a doctoral candidate at State University of New York at Binghamton. Michael Davidson, M.D., is Associate Professor and Director of the Division of Clinical Research, Mt. Sinai School of Medicine. Philio D. Harvev. Ph.D.. is Associate Professor, Director of Psychology Training, Mt. Sinai School of Medicine, and Codrdinator’of Psychiatric CRC, Mt. Sinai Hospital, Mt. Sinai School of Medicine. (Reprint requests to Dr. P.D. Harvey, Dept. of Psychiatry, Box 1229, Mt. Sinai School of Medicine, One Gustave L. Levy Place, New York. NY 10029-6574. USA.) 0165-1781,911$03.50

@ 1991 Elsevier Scientific

Publishers

Ireland

Ltd.

48 Perhaps the most extensively investigated cognitive dysfunction in schizophrenia is the inability to sustain vigilance in attention-demanding situations (e.g., Shakow, 1967; Nuechterlein et al., 1986). The Continuous Performance Test (CPT) may be the most widely used laboratory measure of vigilance in schizophrenic patients (Cornblatt et al., 1988, 1989). The CPT involves monitoring a continuous series of rapidly presented stimuli. Subjects are typically required to press a response key each time a preidentified target item or sequence occurs. Schizophrenic patients, schizophrenic patients in remission, and children at risk for the disorder have manifested deficits on the CPT with a fair degree of consistency (e.g., Asarnow et al., 1977; Kornetsky and Orzack, 1978; Nuechterlein et al., 1986; Cornblatt and Erlenmeyer-Kimling, 1985). Numerous variations of the CPT have been developed over the years. Some versions of the task used target sequences composed of stimuli that would require minimal effort to commit to memory. For example, the early CPT versions only required subjects to press a response key every time the letters “A” followed by “X” appeared (Wohlberg and Kornetsky, 1973). Later versions used series of playing cards (Cornblatt and Erlenmeyer-Kimling, 1985) digit sequences (e.g., 6-4; Buchsbaum et al., 1985), or spatial patterns (Cornblatt et al., 1988), leading to target sequences that are more complex and conceivably more difficult to keep continuously in mind. Other manipulations that were designed to augment the difficulty of the task used stimuli that were distorted in various ways (Nuechterlein et al., 1986). Although all versions of the CPT assess some aspects of vigilance, they are presumed to vary in the amount of information-processing resources required (Nuechterlein et al., 1986). Deficits in the ability to perform information-processing operations requiring controlled (i.e., serial, resource-limited) processes have been suggested to be the most significant cognitive deficits in schizophrenia (Neale and Oltmanns, 1980; Callaway and Nagdhi, 1982), underscoring the importance of developing an understanding of the effects of varying the processing requirements for the CPT. In studying this issue, Cornblatt et al. (1988, 1989) manipulated memory demands for CPT target information and found that increased memory load led to a greater decrease in the performance of schizophrenic patients than in that of normal subjects. Serper et al. (1990) directly examined the extent to which the CPT required controlled processing resources by asking subjects to complete the CPT while simultaneously performing an attention-demanding secondary task. Dual task demands resulted in a notable deterioration in CPT performance relative to performing the central task alone, confirming the suggestions by Nuechterlein et al. (1986) that the CPT is a controlled information-processing task. It has been suggested that CPT versions using overlearned stimuli (e.g., target letters or numbers) require fewer processing resources than tasks containing unfamiliar stimuli with multiple features (e.g., spatial patterns or degraded stimuli; Nuechterlein et al., 1986). Previous studies that examined stimulus complexity as a factor in the CPT compared the relative abilities to process spatial versus verbal stimuli (Cornblatt et al., 1989) or distorted or nondistorted images (Nuechterlein et al., 1986) but did not examine familiarity manipulations within a single type of stimuli (i.e., within either spatial or verbal modalities).* The present study attempts to clarify the role of stimulus familiarity on CPT

49 performance by testing subjects on the CPT using either lexical (3-letter words) or nonlexical verbal stimuli (CVC nonsense syllables). If the hypothesis of Nuechterlein et al. (1986) is correct, the lexical task version, using overlearned test stimuli, will result in generally better task performance than the nonlexical task version and might also be expected to induce a relatively greater performance deficit within schizophrenic patients compared to normal subjects. Also, visual and auditory versions of the CPT were created to examine whether CPT performance parameters were affected by the stimulus modality of the target stimuli. This issue has generally been ignored in research on vigilance deficits in schizophrenic patients, and construct validity for the notion of vigilance deficits would accrue if there were no stimulus modality effects. In addition, the present study examines to what extent medication status influences patients’ vigilance performance. Only a few previous studies have looked at this issue. Orzack et al. (1967) and Spohn et al. (1977), examining the effect of treatment on CPT performance, found that errors of omission and commission were reduced over the course of neuroleptic treatment. Similarly Serper et al. (1990) found that only medicated schizophrenic patients benefited from repeated practice with the CPT. They suggested that medication increases the ability of schizophrenic patients to benefit from practice by enhancing their attentional capacities. Varying the CPT’s processing load through manipulation of the familiarity of the target information may better reveal the influences of medication on schizophrenic patients’ vigilance performance. It is possible that medication might directly affect CPT performance by improving attentional capacity. Alternatively, medication might indirectly improve CPT performance by reducing interfering symptoms such as motor deficits (Walker and Green, 1982). Ratings of thought disorder and affective flattening were collected on the patients to examine differences in clinical symptoms between medicated and unmedicated patients. Methods Subjects. Subjects were 17 unmedicated and 17 medicated schizophrenic inpatients at Mount Sinai Medical Center and the Bronx VA Medical Center, and 19 normal control subjects recruited from the clerical and maintenance staff at a developmental center and from students at a university. Normal control subjects were screened for the presence of personal or familial psychopathology by trained interviewers. Diagnostic information for the hospitalized subjects was obtained by a concurrent interview with two trained interviewers who used the Schedule for Affective Disorders and Schizophrenia (SADS; Spitzer et al., 1978). Information from the diagnostic interview was combined with information collected from the subjects’ charts, and was used by the interviewers to generate independent DSM-III diagnoses (American Psychiatric Association, 1980). In no instances did the charts constitute the sole source of diagnostic information. These independent diagnoses were presented at a consensus meeting with a senior clinician who determined the final diagnosis in conjunction with the two interviewers. Diagnostic reliability was quite high: 0.87 (K), based on a full sample of 187 subjects. Subjects with a hospital diagnosis of a disorder of the central nervous system, substance abuse, or mental retardation were not approached to participate. To be considered for participation, a schizophrenic patient had to meet the following criteria: (1). Drug-free for 3 weeks at the time of admission, due to discontinuation of medication or no history of medication treatment; or (2) on medication and willing to be experimentally discontinued from medication for at least 3 weeks and then returned to

50 treatment. Thus, all patients had to be available for assessment in either medication state; it was then randomly determined as to when patients’ would be tested-either on medication or off medication. All patients tested on medication received a standard treatment of haloperidol, 20 mg, and benztropine, 2 mg, for 2 weeks before testing. Two patients assigned to off-medication testing were placed on medication before they could be examined. These patients were not reassigned to the on-medication group and their data have been excluded from all analyses. There were no subjects who were unable to complete the assessment procedure. Table 1 presents demographic data for the three groups. Analyses of variance (ANOVAs) were used to compare the three groups, who did not differ on any demographic variable except for education (p < 0.05) with the control group being better educated than the schizophrenic groups as determined by Tukey tests.

Table 1. Descriptive information on subjects GrOUD

Normal controls Variable Age Education

(yr)

% Female No. of prior admissions Age at 1st admission

Medicated schizophrenics

Unmedicated schizophrenics

Mean

SD

Mean

SD

Mean

SD

28.40

8.32

31.64

8.80

31.12

7.12

13.80

1.88

12.05

1.34

12.47

1.70

47

41

41

4.12

3.00

3.94

2.75

20.24

4.93

20.88

6.96

Continuous Performance Test Stimuli. The lexical stimuli for the auditory and visual CPTs consisted of 20 three-letter words. These words were affectively neutral based on a criterion of -0.50 < z < 0.50 on the Evaluation (good-bad) continuum of the semantic differential (Osgood et al., 1957) according to the Heise (1965) word norms. Two different IO-word lists were used, with the order of assignment to auditory or visual conditions counterbalanced across subjects. The target sequence for List 1 consisted of the word “DAY” followed by the word “FAR,” and the target sequence for List 2 consisted of the word “SIT” followed by the word “PAY.” The nonlexical stimuli for the CPTs consisted of 20 CVC nonsense syllables. As with the lexical stimuli, two IO-item stimulus lists were used. The target sequence for List 1 consisted of “GOK” followed by “SIJ,” and for List 2, the target sequence consisted of the nonsense syllable “CUG” followed by “DOF.” The visual stimuli were presented for a duration of 500 ms with a fixed interstimulus interval (ISI) of 1000 ms. Each visual stimulus was presented on a monochromatic computer monitor and was 5.5 cm wide X 1.5 cm high. The visual stimuli were rapidly faded after presentation. The auditory stimuli were presented over headphones in an audiotaped format in a female voice, with an IS1 of 1000 ms. Apparatus. For the visual CPTs, response timing began automatically upon presentation of the stimulus. For the auditory CPT, the onset of each spoken stimulus word was detected by the computer, and when onset was detected, response timing began. Research assistants monitored a control box during the testing procedure to ensure that the subject was not keeping the button pressed throughout or failing to press the button with enough pressure to error for produce a detectable response. For the visual CPTs, the maximum presentation stimulus duration or IS1 was f 2.5 ms. The maximum error for the recorded reaction times to all stimuli was + 5.0 ms. The reliability of the timing was determined by the system crystal oscillator.

51 Motor Proficiency Tasks. Motor proficiency was examined because of previous suggestions (Walker and Green, 1982) that neuromotor variables might account for some aspects of schizophrenic patients’ CPT deficits through interference with response production. The assessment was conducted using two tasks from the Upper-Limb Speed and Dexterity subtest of the Bruininks-Oseretsky Test of Motor Proficiency (Bruininks, 1978). The first task was a simultaneous motor speed task, requiring subjects to place pennies in two boxes with both hands simultaneously, with the dependent variable the time (in set) it took to place the 12 pairs. The other task was a motor speed and dexterity test, asking subjects to displace pegs on a pegboard using their preferred hand. Subjects were given 15 set to displace as many of the 25 pegs as possible. Table 2 lists the motor proficiency scores.

Table 2. Motor performance

and clinical information

on all subjects

Group Normal controls Variable

Medicated schizophrenics

Unmedicated schizophrenics

Mean

SD

Mean

SD

Mean

SD

12.90

1.97

19.94

5.65

21.53

10.08

13.60

1.67

10.53

3.37

9.71

1.99

2.94

2.30

3.76

2.41

1.00

1.17

1.71

1.04

2.00

1.74

2.25

1.40

Motor proficiency data Penny

placing time

No. of pegs displaced

Clinical symptom data Positive thought

disorder

Negative

thought

Affective

flattening

disorder

Clinical Ratings of Thought Disorder and Affective Flattening. Clinical speech disorder ratings were completed by the SADS raters for all subjects from the conversation during the SADS interview. These ratings were based on the criteria in the Scale for the Assessment of Thought, Language, and Communication Disorders (TLC; Andreasen, 1979). Three positive sign speech disorders (Pressure of Speech, Derailment, and Tangentiality) were extracted from the TLC ratings, and two negative sign speech disorders (Poverty of Speech and Poverty of Content) were extracted from the complete set of ratings. The overall reliability of these thought disorder ratings, on the basis of 177 ratings conducted by these raters, was 0.80 (K). Affective flattening ratings were generated by averaging the items from the Scale for the Assessment of Negative Symptoms (SANS; Andreasen, 1981) flat affect subscale. The item score for inappropriate affect was excluded, as were the ratings on subjective impressions and the global rating. These ratings, like the TLC ratings, were generated by the interviewers on the basis of data collected from the structured diagnostic interview and from chart review. The reliability of these ratings, based on 120 interviews, was 0.82 (ICC). Table 2 presents these scores. Procedure. Patients were interviewed with the SADS, and then trained research assistants tested the subjects on the motor proficiency tasks within 2 days of that initial interview. After this assessment, subjects were tested on the CPTs. For the visual tasks, subjects were seated 0.5 m from the monitor. Auditory stimuli were presented over headphones in a tape-recorded format. Subjects were given instructions to monitor the series of words or nonsense syllables presented in the center of the screen, one at a time. They were required to press a hand-held button as quickly as possible, using their preferred hand, following each presentation of the target sequence. A series of 10 practice trials in each condition preceded the testing. In each condition, the target sequence appeared 12 times. The two members of the target sequence also appeared individually, on three occasions each. There were 60 trials in each of the four tasks, and the order of assignment of the stimulus lists and the order of presentation of the four tasks were both counterbalanced.

52

Results Clinical and Motoric Variables. Comparisons by t tests did not reveal any differences between the two schizophrenic samples on any of the three clinical measures of symptom severity. The unmedicated patients scored higher on all three of the clinical measures, however. The two motor competence variables, the time required to sort the pennies and the number of pegs displaced, were compared across the three groups with one-way ANOVAs. The group effect was significant for penny placing time (F= 4.61; df = 2, 51; p < 0.05. Tukey followup tests revealed that the unmedicated patients were significantly slower than the normals, with no other differences being significant. The group effect for the number of pegs displaced was also significant (F = 8.35; df = 2,5 1;p < 0.005). Tukey tests revealed that both schizophrenic groups displaced significantly fewer pegs than the normals. CPT Error Scores. The normal subjects’ performance on the CPT was notably better than that of the schizophrenic patients, and there were very few errors. On the basis of t tests, the normal subjects made significantly fewer errors of omission and commission than the least deviant schizophrenic sample on each of the eight CPT error measures. Thus, the analyses of the CPT errors were performed with two medication status X modality (auditory-visual) X stimulus type (lexical-nonlexical) ANOVAs, one for errors of omission and one for errors of commission. Table 3 presents these CPT errors. Table 3. CPT error scores for all subjects Grow Normal controls Variable

Medicated schizophrenics

Unmedicated schizophrenics

Mean

SD

Mean

SD

Mean

SD

Errors of omission Visual task Lexical Nonlexical Auditory task Lexical Nonlexical

0.00

0.00

1.71

1.90

2.00

1.94

0.26

0.45

1 s9

1.94

1.88

1.62

0.05

0.23

1.53

2.92

1.41

1.70

0.21

0.54

0.59

1.12

2.82

3.45

0.26

0.45

1.82

2.79

2.24

4.58

0.37

0.60

1.53

2.63

5.24

8.22

0.16

0.38

1.12

3.46

2.29

3.37

0.26

0.56

1.23

2.41

7.00

9.97

Errors of commission Visual task Lexical Nonlexical Auditory task Lexical Nonlexical

For errors of omission, the three-way interaction of medication status X modality X stimulus type was nonsignificant, as were the two-way interactions of medication

53 status X modality and modality X type. The two-way interaction of medication status X stimulus type was significant (F= 4.61; df = 1,32; p < 0.05). TO follow up this interaction, the total numbers of lexical and nonlexical errors of omission were summed across the nonsignificant stimulus modality variable and compared across medication status. One-way simple effects tests (Winer, 1972) revealed that unmedicated patients made more nonlexical errors of omission than medicated patients (F= 4.49; df = 1,32; p < 0.05) and that the groups failed to differ on lexical errors of omission (F= 0.02; df = 1, 32, NS). Since the distributions of CPT errors were not normal, these two comparisons were performed with nonparametric statistics. Mann-Whitney U tests confirmed the above results: On nonlexical errors of omission, there was a significant group difference (U = 83.5, p < 0.05) and on lexical errors of omission, there was no significant group difference (U = 126.5, p = 0.53). For errors of commission, the same pattern of results was found. No interactions or main effects involving stimulus modality were significant, but the two-way interaction of medication status X stimulus type was significant (F= 5.32; df = 1,32; p < 0.05). When this interaction was followed up, it was found that unmedicated patients made more errors of commission on nonlexical stimuli (F= 4.92; df = 1,32; p < 0.05), but the groups did not differ in their number of errors of commission with lexical stimuli (F = 0.59; df = 1, 32; NS). When these analyses were repeated nonparametrically, the same results were found: Unmedicated patients made significantly more errors of commission on nonlexical stimuli (U = 68.0, p < 0.01) and the groups did not differ on lexical errors (U = 103.5, p = 0.16). Correlational Analyses. These analyses were oriented toward identification of the correlations between the CPT variables and motor variables in the patient samples. Since the patterns of between-group differences were the same for errors of omission and commission across stimulus modality, composite scores for the total number of lexical CPT errors, nonlexical CPT errors, and total errors were computed and correlated with the motoric variables. Since the two motor competence variables were highly correlated within each of the schizophrenic groups (r > 0.8), the peg displacement score was the only one used in the analyses. The correlations between all three CPT error scores and the motor competence variable were nonsignificant in the unmedicated group (total: r = -0.17; lexical: r = -0.25; nonlexical: r = -0.12). In the medicated sample, all three correlations were large and significant (total: r = -0.70; lexical: r = -0.72; and nonlexical: r = -0.64). All three correlations in the medicated sample exceeded the 95% confidence interval for the correlations in the unmedicated sample, indicating that they were significantly (p < 0.05) larger in magnitude in the medicated sample than in the unmedicated one.1 I. Since the correlation between motor proficiency and CPT errors was consistently different in the two patient samples, analysis of covariance was used to reassess the group differences in lexical and nonlexical errors of omission and commission. When the medication status analyses were performed with the number of pegs displaced as a covariate, the between-group differences remained. The covariate effect of pegs was significant in three of the four analyses, but the pattern of significant group differences on nonlexical errors was preserved.

54

Discussion The present results suggest that schizophrenic patients are quite deficient in the ability to identify two-item target sequences relative to normal subjects across a variety of stimulus modality and lexicality conditions, with unmedicated patients showing a specific deficit in the ability to identify nonlexical target sequences. The stimulus modality of the information did not affect the performance of any subjects, suggesting that the processes involved in the CPT are similar regardless of whether target information is presented visually or auditorally. Similar to the results of Cornblatt et al. (1989) and Nuechterlein et al. (1986), a manipulation designed to increase the processing load of schizophrenic patients led to a decrease in performance. In the present study, however, this manipulation led to a specific deficit in performance on the part of patients who were medication free. The Nuechterlein et al. (1986) study demonstrated that increasing the difficulty of the stimulus discrimination augmented the difficulty of the task, in a largely medicated sample. The Cornblatt et al. (1989) study demonstrated, also in medicated samples of schizophrenic patients, that increasing processing load by increasing the difficulty of retention of the target sequence led to relatively greater decreases in schizophrenic patients’ ability to perform the CPT, as compared to normal subjects. It would be of interest to determine the extent to which medication affects the relatively greater deterioration in performance shown by schizophrenic patients with those two difficulty manipulations. The present data underscore the possibility that CPT performance deficits may be due to difficulties in the ability to remember the target sequence while monitoring the stream of potential target stimuli. A nonsense syllable sequence is likely to be more difficult than a word sequence to retain in active memory while performing the attention-demanding task of monitoring the sequence of potential target stimuli because it is constituted of unfamiliar information. Nonsense syllables may be more difficult to discriminate from each other than three-letter words because of the lower level of familiarity of nonsense syllables relative to three-letter words. Since the CPT is a controlled processing task, increases in perceptual load alone may make the task of recalling the target sequence more difficult because of resource limitations, leading to an interactive effect. The fact that medication is associated with improved performance on the nonlexical conditions of the present task can be used to suggest that the source of the performance deficit is in the cognitive operations required to retain the target sequence in active memory while performing stimulus-monitoring operations, instead of in the perceptual operations associated with rapid discrimination of the targets. Earlier studies (Oltmanns et al., 1979; Harvey and Pedley, 1989) have suggested that medication in schizophrenic patients improves performance on tasks that require controlled information processing but has relatively little impact on tasks measuring basic perceptual operations (Killian et al., 1984). Thus, the improved performance of medicated patients on the nonlexical condition probably reflects an enhanced ability to perform multiple attention-demanding operations concurrently, rather than a perceptual advantage. In line with our interpretations of the results of a recent study (Serper et al., 1990), it appears that neuroleptic medica-

55 tion enhances attentional capacity in schizophrenic patients, although it does not lead to normal-range performance without extensive practice. The motoric response components of the task are not responsible for CPT deficits in all schizophrenic patients, as evidenced by the lack of correlation between motor and CPT variables in the unmedicated patients. The variance in the correlations of CPT errors and motor variables across medication status underscores the importance of assessment of unmedicated patients, however, in order to understand the cognitive functioning of schizophrenic patients both on and off medication. It should be noted that these motor skills tests do have, at least, a focused attention component. As a consequence, they are not attention-free estimates of motor skills and cannot be seen to be completely independent attentional abilities that might determine CPT performance. The primary limitation of the study is its use of a cross-sectional rather than a retest design. As a result, it cannot be determined whether treatment response would affect the results. A study now in process examines a variety of attentional variables during treatment and may resolve some of the limitations of the present study. Acknowledgment. This research was presented in part by the first author as a doctoral dissertation to the Department of Psychology at SUNY at Binghamton. The authors thank the patients and nursing staff on the KCC 8 South and Bronx VA 3B2 clinical research units.

References American Psychiatric Association. DSM-III: Diagnostic and Statistical Manual of Mental Disorders. 3rd ed. Washington, DC: APA, 1980. Andreasen, N.C. The Scale for Assessment of Thought, Language, and Communication: I. Clinical assessment, definition of terms, and assessment of their reliability. Archives of General Psychiatry, 36:1315-1321, 1979. Andreasen, N.C. The Scale for the Assessment of Negative Symptoms. Unpublished Manual, University of Iowa, Iowa City, IA, 1981. Asarnow, R.F., and MacCrimmon, D. Residual performance deficit in clinically remitted schizophrenics: A marker of schizophrenia? Journal of Abnormal Psychology, 87597-608, 1978. Asarnow, R.F.; Steffy, R.; Cleghorn, J.M.; and MacCrimmon, D.J. An attentional assessment of foster children at risk for schizophrenia. Journal of Abnormal Psychology, 86: 267-275, 1977. Bleuler, E. Dementia Praecox or the Group of Schizophrenias. New York: International Universities Press, l911/ 1950. Bruininks, R.H. Bruininks-Oseretsky Test of Motor Proficiency. Circle Pines, MN: American Guidance Service, 1978. Buchsbaum, M.S.; Haier, R.J.; Sostek, A.J.; Weingartner, H.; Zahn, T.P.; Siever, L.J.; Murphy, D.L.; and Brody, L. Attentional dysfunctions and psychopathology in college men. Archives of General Psychiatry, 42:354-360, 1985. Callaway, E., and Naghdi, S. An information-processing model for schizophrenia. Archives of General Psychiatry, 39:339-347, 1982. Cornblatt, B., and Erlenmeyer-Kimling, L. Global attentional deviance as a marker of risk for schizophrenia: Specificity and predictive validity. Journal of Abnormal Psychology, 94:470-486, 1985. Cornblatt, B.A.; Lenzenweger, M.F.; and Erlenmeyer-Kimling, L. The Continuous Performance Test, Identical Pairs version: II. Contrasting attentional profiles in schizophrenic and depressed patients. Psychiatry Research, 29:65-85, 1989.

56 Cornblatt, B.A.; Risch, N.J.; Faris, G.; Friedman, D.; and Erlenmeyer-Kimling, L. The Continuous Performance Test, Identical Pairs version (CPT-IP): I. New findings about sustained attention in normal families. Psychiatry Research, 26:223-238, 1988. Harvey, P.D.; Docherty, N.M.; Serper, M.R.; and Rasmussen, M. Cognitive deficits and thought disorder: II. An eight month followup study. Schizophrenia Bulletin, 16:147-156, 1990. Harvey, P.D., and Pedley, M. Auditory and visual distractibility in schizophrenia: Clinical and medication status correlations. Schizophrenia Research, 2:295-300, 1989. Heise, D.R. Semantic differential profiles for 1,000 most frequent English words. Psychological Monographs, 79: Whole No. 601, 1965. Killian, G.A,; Holzman, P.S.; Davis, J.M.; and Gibbons, R. Effects of psychotropic medication on selected cognitive and perceptual measures. Journal of Abnormal Psychology, 9358-70, 1984. Kornetsky, C., and Orzack, M.H. Physiological and behavioral correlates of attention in schizophrenic patients. In: Wynne, L.C.; Cromwell, R.L.; and Matthysse, S., eds. The Nature ofSchizophrenia. New York: Wiley, 1978. pp. 196-204. Kraepelin, E. Dementia Praecox and Paraphrenia. Edinburgh: Livingstone, 1919. Neale, J.M.; McIntyre, C.W.; Fox, R.; and Cromwell, R.C. Span of apprehension in acute schizophrenics. Journal of Abnormal Psychology, 74:593-596, 1969. Neale, J.M., and Oltmanns, T.F. Schizophrenia. New York: Wiley, 1980. Nuechterlein, K.H.; Edell, W.S.; Norris, M.; and Dawson, M.E. Attentional vulnerability indicators, thought disorder, and negative symptoms. Schizophrenia Bulletin, 12:408-426, 1986. Oltmanns, T.F. Selective attention in manic and schizophrenic psychoses: The effect of distraction on information processing. Journal of Abnormal Psychology, 87:212-225, 1978. Oltmanns, T.F.; Ohayon, J.; and Neale, J.M. The effect of medication and diagnostic criteria on distractibility in schizophrenia. Journal of Psychiatric Research, 14:81-91, 1979. Orzack, M.H.; Kornetsky, C.; and Freeman, H. The effects of daily administration of carphenazine on attention in the schizophrenic patients. Psychopharmacologia, 11:31-38, 1967. Osgood, C.E.; Suci, G.J.; and Tannenbaum, P.H. The Measurement of Meaning. Urbana, IL: University of Illinois, 1957. Rund, B.R. The effects of distraction on focal attention in paranoid and nonparanoid schizophrenic patients compared to normals and nonpsychotic psychiatric patients. Journal of Psychiatric Research, 17:241-250, 1983. Saccuzzo, D.P.; Hirt, M.; and Spencer, T.J. Backward masking as a measure of attention in schizophrenia. Journal of Abnormal Psychology, 83:512-522, 1974 Serper, M.R.; Bergman, R.L.; and Harvey, P.D. Medication may be required for the development of automatic information processing in schizophrenia. Psychiatry Research, 32:281-288, 1990. Shakow, D. Segmental set: A theory of formal psychological deficit in schizophrenia. Archives of General Psychiatry, 6: 17-33, 1967. Spitzer, R.L.; Endicott, J.; and Robins, L. The Schedule for Affective Disorders and Schizophrenia. New York: Biometrics Research, 1978. Spohn, H.E.; Lacoursiere, R.B.; Thompson, K.; and Coyle, L. Phenothiazine effects on psychological and psychophysiological functions in chronic schizophrenics. Archives of General Psychiatry, 341633-644, 1977. Walker, E., and Green, M.F. Motor proficiency and attentional-task performance by psychotic patients. Journal of Abnormal Psychology, 91:261-268, 1982. Winer, B.J. Statistical Principles in Experimental Design. 2nd ed. New York: McGraw-Hill, 1972. Wohlberg, G.W., and Kornetsky, C. Sustained attention in remitted schizophrenics. Archives of General Psychiatry, 28:533-537, 1973.