Inhibitory control and spatial working memory: A saccadic eye movement study of negative symptoms in schizophrenia
Article Outline
Abstract
The negative symptoms of schizophrenia are perhaps the most unremitting and burdensome features of the disorder. Negative symptoms have been associated with distinct motor, cognitive and neuropathological impairments, possibly stemming from prefrontal dysfunction. Eye movement paradigms can be used to investigate basic sensorimotor functions, as well as higher order cognitive aspects of motor control such as inhibition and spatial working memory — functions subserved by the prefrontal cortex. This study investigated inhibitory control and spatial working memory in the saccadic system of 21 patients with schizophrenia (10 with high negative symptoms scores and 11 with low negative symptom scores) and 14 healthy controls. Tasks explored suppression of reflexive saccades during qualitatively different tasks, the generation of express and anticipatory saccades, and the ability to respond to occasional, unpredictable (“oddball”) targets that occurred during a sequence of well-learned, reciprocating saccades between horizontal targets. Spatial working memory was assessed using a single and a two-step memory-guided task (involving a visually-guided saccade during the delay period). Results indicated significant increases in response suppression errors, as well as increased response selection impairments, during the oddball task, in schizophrenia patients with prominent negative symptoms. The variability of memory-guided saccade accuracy was also increased in patients with prominent negative symptom scores. Collectively, these findings provide further support for the proposed association between prefrontal dysfunction and negative symptoms.
Keywords: Schizophrenia, Negative symptoms, Saccades, Inhibition, Spatial memory
1. Introduction
The negative symptoms of schizophrenia are perhaps the most unremitting and burdensome features of the disorder, interfering significantly with activities of daily living (Andreasen and Olsen, 1982). Although first described early last century by Kraepelin (1919/1971), the characterization of negative symptoms remains unclear. Factor analytical studies have most strongly supported the inclusion of apathy, flat affect and isolation (Peralta et al., 1992). Avolition, psychomotor retardation, anhedonia and attentional impairment are specifically less associated with negative symptoms, but included in some rating scales (Peralta and Cuesta, 1996).
Patients with prominent negative symptoms have been shown to have motor (Caligiuri et al., 1998, Jogems-Kosterman et al., 2001), cognitive (Hammer et al., 1995, Berman et al., 1997) and neuropathological differences (Rodríguez et al., 1997, Sanfilipo et al., 2000, Wible et al., 2001), when compared to patients who do not have prominent negative symptoms. Structural brain imaging investigations have suggested cortical–striatal–thalamocortical circuit dysfunction including predominantly prefrontal cortical regions as the neural basis for primary and enduring negative symptoms (Rodríguez et al., 1997, Sanfilipo et al., 2000, Wible et al., 2001). The results of neuropsychological studies also support the involvement of prefrontal dysfunction (Hammer et al., 1995, Berman et al., 1997).
Inhibitory control and spatial working memory are two interrelated functions that rely on the prefrontal cortex (Fuster, 1985, Fuster, 1997, Goldman-Rakic, 1987). Eye movement tasks offer a useful and sensitive behavioural tool to investigate both inhibitory control and spatial working memory. The neural systems controlling saccades (rapid eye movements) have been extensively investigated, permitting insights into patterns of abnormality that can provide important clues about location of neuropathology (for reviews see Pierrot-Deseilligny et al., 1995, Leigh and Zee, 1999).
Previous saccade studies exploring inhibitory control and spatial working memory in schizophrenia have yielded somewhat discrepant findings, perhaps partially relating to population sampling differences in this heterogeneous condition (see Broerse et al., 2001 for a review). Antisaccades, generated to the mirror location (i.e. same amplitude, opposite direction) of a visible target, are commonly used to explore inhibitory control. Saccades incorrectly generated toward the visible target are classified as inhibition errors. Of the few saccade studies exploring the relationship between these measures and negative symptoms, the most prominent findings indicate increased inhibition errors during the antisaccade task, particularly in those patients with more severe negative symptoms (Tien et al., 1996, Nkam et al., 2001, Ettinger et al., 2004, Ettinger et al., 2006).
Deficiencies in inhibitory control can manifest in a number of ways. Increases in the occurrences of anticipatory saccades (generated before the appearance of forthcoming, predictable targets) and express (very fast) saccades (Fischer and Ramsperger, 1984) have been previously reported in schizophrenia patients and may represent impaired inhibitory control (McDowell et al., 1996, McDowell et al., 1997, Clementz, 1998, Karoumi et al., 1998, Spengler et al., 2006). Although one study reported a positive correlation between the percentage of anticipatory saccades and negative symptom scores (Karoumi et al., 1998), these tasks have not been previously explored specifically in those patients with prominent negative symptoms.
Another aspect of inhibitory control involves the suppression of expected, but no longer appropriate movements. “Oddball tasks” are those where participants perform a sequence of well-learned, reciprocating saccades and are occasionally required to inhibit a saccade to an expected location and instead generate a saccade to an unpredictable ‘oddball’ target (Winograd-Gurvich et al., 2006). This form of saccadic inhibition has not been assessed in schizophrenia.
The primary aim of this study was to explore the relationship between negative symptoms in schizophrenia and performance on eye movement tasks relying on the integrity of the prefrontal cortex. To extend previous suggestions of deficient inhibition in negative symptoms, inhibitory control was explored using a variety of saccade tasks that required response suppression and response selection. Spatial working memory, a second function associated with prefrontal functioning, was also investigated in two tasks: a single memory-guided saccade task and a more challenging two-step memory-guided task. Given the proposed prefrontal dysfunction in the negative symptom group, we hypothesized that inhibitory deficiencies, as well as spatial working memory deficits, would be more pronounced in those patients with high negative symptom scores, as compared to patients with low negative symptom scores.
2. Method
2.1. Subjects
Twenty-one individuals (11 males, 10 females) meeting DSM-IV diagnostic criteria for schizophrenia (American Psychiatric Association, 2000), with disease duration ranging from 1 to 40 years, participated. All patients were taking antipsychotic and/or adjuvant medications (Table 1). A comparison group of 14 neurologically healthy, control subjects (4 males, 10 females) with no history of affective or psychotic illness also participated.
Table 1. Clinical data for schizophrenia patients
| Low NS group | High NS group | ||||
|---|---|---|---|---|---|
| Gender, age (years) | Disease duration (years) | Medication and daily dosage (mg) | Gender, age (years) | Disease duration (years) | Medication and daily dosage (mg) |
| M, 39 | 11 | Quetiapine 400 | M, 40 | 18 | Olanzapine 12.5 |
| Diazepam | Sodium valproate 300 | ||||
| M, 25 | 5 | Amisulpride 600 | F, 57 | 40 | Clozapine 350 |
| Sodium valproate 1500 | |||||
| F, 33 | 7 | Paroxetine 20 | M, 32 | 3 | Clozapine 400 |
| Olanzapine 130 | Sodium valproate (2 tables) | ||||
| Sodium valproate 100 | Venlafaxine | ||||
| M, 27 | 6 | Olanzapine 10 | M, 30 | 8 | olanzapine 15 |
| F, 28 | 6 | Clozapine (3.5 tablets) | M, 32 | 3 | Aripiprazole 30 |
| Amisulpride (2 tablets) | Diazepam (as needed) | ||||
| Sodium valproate | |||||
| F, 54 | 12 | Aripiprazole 15 | F, 43 | 11 | Sertraline 300 |
| Clozapine 400 | |||||
| Sodium valproate 500 | |||||
| M, 36 | 4 | Olanzapine 10 | M, 30 | 6 | Risperidone depot |
| M, 47 | 28 | Aripiprazole 30 | F, 54 | 15 | Olanzapine 100 |
| Amisulpride | Quetiapine 75 | ||||
| Carbamazepine 1200 | |||||
| F, 39 | 13 | Olanzapine 15 | F, 51 | 1 | Risperidone 2 |
| Venlafaxine 150 | |||||
| M, 39 | 13 | Amisulpride 400 | F, 49 | 13 | Olanzapine 15 |
| F, 58 | 19 | Amisulpride 400 | |||
To reduce confounding effects, patients were selected on the following criteria: 1) at the time of assessment all patients were outpatients; 2) patients were in a stable phase of their disease (i.e. medication stabilized for at least 2 weeks prior to assessment; showing no clear florid symptoms); 3) patients were only taking atypical antipsychotics; 4) patients were not diagnosed with other psychiatric or neurological disorders (including stroke and head injury); 5) patients were not showing any clinical signs of drug induced extrapyramidal disorders, and 6) all participants scored above 23/30 on the Mini-Mental State Examination; MMSE (Folstein et al., 1975). In addition, all participants were free from current psychoactive substance use disorders and no participant demonstrated visual impairment (including colour blindness), other than refractive error, or clinical ocular motor pathology.
After complete description of the study, written informed consent was obtained by all participants and ethics approval was given according to the NH and MRC criteria by local ethics committees.
2.2. Clinical assessment
For the patient group, symptoms were assessed using the Positive and Negative Syndrome Scale (PANSS; Kay et al., 1987). Patients were divided into two subgroups with either low negative symptom (Low NS) scores (no more than one or two ‘moderate’ scores out of seven) or high negative symptom (High NS) scores (at least three ‘moderate’ scores) [criteria according to PANSS manual (Kay et al., 1987)]. The Scale for the Assessment of Negative Symptoms (SANS; Andreasen, 1984) was additionally administered for correlation purposes. The presence of extrapyramidal symptoms was evaluated using the Extra Pyramidal Symptoms Scale (EPS; Simpson and Angus, 1970) and the Abnormal Involuntary Movement Scale (AIMS; Guy, 1976). Depressive symptoms were evaluated using the Montgomery and Åsberg Depression Rating Scale (MADRS; Montgomery and Åsberg, 1979). Estimates of premorbid IQ were attained using the National Adult Reading Test (NART; Nelson, 1982) (see Table 2).
Table 2. Mean values and standard deviations (in parentheses) for age and clinical rating scores for patients with high and low negative symptom scores (NS) and controls
| High NS | Low NS | Controls | Statistics | ||||
|---|---|---|---|---|---|---|---|
| Age | 41.8 | (10.50) | 38.64 | (10.74) | 41.8 | (11.48) | F(2,34) =2.81, P=0.76 |
| PANSS | |||||||
| Positive symptoms | 13.1 | (4.79) | 9.09 | (3.02) | F(1,20)=5.37, P=0.03 | ||
| Negative symptoms | 16.7 | (2.45) | 11.09 | (2.98) | F=(1,20)=21.90, P=0.000 | ||
| General psychopathology | 28.3 | (5.31) | 21.64 | (4.39) | F=(1,20)=9.89, P=0.005 | ||
| SANS — total score | 26.7 | (11.56) | 7.55 | (5.75) | F=(1,20)=23.82, P=0.000 | ||
| EPS | 2.5 | (1.72) | 0.64 | (1.29) | F=(1,20)=2.11, P=0.16 | ||
| AIMS | 1.3 | (1.35) | 0.64 | (1.29) | F=(1,20)=2.81, P=0.11 | ||
| MADRS | 16.4 | (12.60)⁎⁎ | 11.82 | (8.92)⁎⁎ | 0.58 | (1.38) | F=(2,32)=9.92, P=0.00 |
| NART | 113.08 | (13.95) | 108.44 | (9.57) | 116.68 | (10.38) | F=(2,30)=1.45, P=0.00 |
2.3. Eye movement recordings and apparatus
Horizontal eye movements were recorded using an infrared eye tracking system (Skalar, IRIS), with output sampled at 1 kHz. Output from the eye tracker was displayed on the analysis computer screen alongside a control signal generated by E-Prime which indicated stimulus change. To avoid any time lags due to screen refresh rates, a photodiode was placed directly over a non-visible portion of the stimulus screen to concurrently record stimulus change in real time. Participants were seated in semi-darkness, 55 cm from a flat LCD screen where green target crosses measuring 17 mm×17 mm were displayed along a horizontal axis at 2.5° intervals from 15° left to 15° right. The subject's head was stabilized by a chin rest and bite bar. Each experimental task was preceded by a calibration sequence in which subjects fixated three green crosses located along the horizontal axis (−10°, 0°, +10°). Saccade tasks were presented in the following order and there were rest breaks between all protocols. Instructions were provided prior to each task and slow-paced practice trials were presented before more challenging tasks. All participants completed all tasks; however, the data from one participant in the high NS group during the reflexive saccade task were excluded due to excessive blinking.
2.4. Saccade tasks
The visually-guided reflexive saccade task involved an unpredictable sequence of green target crosses presented on a horizontal axis (with targets located at either −10°, −5°, 0° [centre], +5°, +10°). Each green cross was illuminated for 2, 4 or 6 s in a pseudorandom order, and as one green cross extinguished, the next green cross immediately illuminated. Target sequence was unpredictable in terms of direction (right or left) and amplitude (5°, 10°, or 15°); see Fig. 1a. There were 16 trials for each amplitude.
Fig. 1.
Schematic representations of saccade tasks showing stimuli on LCD screen alongside a time-line: (a) reflexive, visually guided task; (b) oddball task; (c) single memory-guided task; and (d) two-step memory-guided task.
A gap paradigm was included to explore express saccades. Two variations to the above protocol were that each target was illuminated for 2 s, and a gap (i.e. blank screen) of either 200 ms or 250 ms was presented between each target. There were 15 trials for each gap interval. Participants were instructed to fixate the centre of each green cross and not move their eyes until the next cross appeared.
The predictive paradigm involved a green target cross alternating between fixed positions (+5° and −5°) at a fixed time frequency (1 target/s) for 1 min (i.e. 60 trials). Therefore, task requirements were fully predictable in terms of time and space and allowed the exploration of anticipatory saccades.
The oddball task investigated the ability to inhibit an expected motor program and reprogram a new saccade to correspond to an “oddball” target (see Fig. 1b). Green target crosses alternated between two positions (7.5° left and right of centre) and remained illuminated for 2 s with no gaps between targets. On 15 pseudorandom occasions, an oddball target appeared, in the opposite direction to the expected target location, requiring on-line reprogramming. For the purpose of this study, only errors were calculated (i.e. saccades made to the expected, rather than oddball, location).
During the single memory-guided task (see Fig. 1c) participants were instructed to fixate the central green cross. The fixation target was displayed for 5 s before a yellow cross was simultaneously flashed for 1000 ms in one of four positions (−10°, −5°, +5°, +10°). Following a delay period of 2 s the green fixation cross extinguished, that acting as the ‘go’ signal to initiate a memory-guided eye movement. After 2 s of darkness a green cross appeared in the position where participants were supposed to be fixating to provide feedback on the accuracy of their memory-guided saccade. There were 15 trials and participants were instructed to fixate the central green cross and not to move their eyes to the flashed yellow cross. When the green cross went out, they were to quickly fixate the remembered position of the yellow cross.
The two-step memory-guided task included one variation to the standard procedure described above with an intervening visually guided saccade in response to a second green target cross that appeared during the delay period (see Fig. 1d). Participants were additionally instructed to follow the green cross as it moved and then when the second green cross extinguished, to look to the remembered position of the yellow cross.
In the fixation-distractor task a central green target cross was illuminated for 60 s. There were no distractions for the first 15 s to ensure basic fixation was intact. The distractor was a yellow cross that appeared randomly on 12 occasions at either −10°, −5°, +10° or +5° while the central fixation remained illuminated. There were an equal number of left and right distractors and each distractor remained flashing for 2 s.
In the saccade-engagement-distractor task a green target cross alternated between two positions (5° left and right of centre) remaining illuminated for 2 s, and as one target extinguished, the next target illuminated. As in the previous task, a yellow distractor cross appeared randomly on 12 occasions to the left or right of the central fixation point and remained flashing for 2 s.
2.5. Data analysis
Eye movements were analysed off-line, using an interactive computer program (Matlab). The beginning and end of a saccade were determined visually, in reference to a position trace and confirmed according to changes in the acceleration of the velocity profile of the saccade trace. Temporal measures considered were latency, defined as the time between target onset and the commencement of the primary saccade and peak velocity of the initial saccade, determined by computer differentiation of the position trace. Accuracy measures included the gain of the primary saccade (the ratio of primary saccade amplitude to target step amplitude) and the gain of the final eye position (the ratio of final eye position amplitude to target step amplitude) where a gain of 1 is desirable. Errors during the memory-guided, fixation and saccade-engagement tasks were defined as a saccade initiated toward the yellow peripheral cross. During the oddball task, responses were classified as errors if the direction of the initial saccade was incorrect (i.e. toward the expected target location, with amplitudes of at least 2°, rather than the oddball target location). Error types were labeled as occurring: (i) before target presentation or (ii) after target presentation (at least 100 ms after presentation of the oddball target). The percentage of anticipatory saccades [saccades occurring prior to, or within 100 ms of target presentation (Karoumi et al., 1998)] and the percentage of express saccades [saccades occurring between 80 and 120 ms of target presentation (Fischer and Ramsperger, 1984)] were also derived for relevant protocols.
In cases where a Levene's test for equality of variance indicated no significant difference between the variances of the three populations (note that alpha was set at 0.01 to examine homogeneity of variance), statistical comparisons were based on Analyses of Variance (ANOVAs). A series of Pearson's product moment correlations were also performed to explore associations between performance on eye movement tasks and scores from clinical ratings (PANSS and SANS ratings). Given that the memory-guided tasks were included to provide information on accuracy (spatial working memory) and inhibitory control (error rates), data relating to temporal measures (latency and velocity) for both single and two-step memory-guided tasks were pooled together and compared to reflexive saccade latencies (Group×Task Two-way ANOVA).
3. Results
3.1. Express and anticipatory saccades
See Table 3 for the percentage of express and anticipatory saccades, as well as means and standard deviations for accuracy and velocity data.
Table 3. Mean values and standard deviations for percentages of express and anticipatory saccades during the gap and predictive paradigms, respectively, as well as accuracy data and peak velocity data for each group
| Ocular motor paradigm | Control | High NS | Low NS |
|---|---|---|---|
| Mean±S.D. | Mean±S.D. | Mean±S.D. | |
| Express saccades | |||
| Percentage (200 ms gaps) | 9.52±7.91 | 36.67±17.65⁎⁎ | 28.79±19.85⁎⁎ |
| Percentage (250 ms gaps) | 19.17±16.87 | 29.47±15.54 | 30.62±13.88 |
| Accuracy — primary saccade (gain) | 1.09±0.39 | 0.92±0.24 | 0.94±0.11 |
| Accuracy — final eye position (gain) | 1.04±0.18 | 0.97±0.24 | 1.06±0.10 |
| Peak velocity (degrees/s) | 361.30±120.68 | 389.14±89.30 | 330.57±58.17 |
| Anticipatory | |||
| Percentage | 72.56±12.11 | 76.00±14.10 | 68.75±23.00 |
| Accuracy — primary saccade (gain) | 0.89±0.11 | 0.82±0.23 | 0.75±0.14 |
| Accuracy — final eye position (gain) | 1.03±0.12 | 0.98±0.17 | 1.00±0.06 |
| Peak velocity (degrees/s) | 308.09±66.87 | 347.19±100.06 | 275.98±81.29 |
The percentage of express saccades for gaps of 200 ms differed significantly between the groups: F(2,34)=10.11, P<0.001. Post hoc analyses revealed a significant difference between the control and low NS group (P<0.01) and between the control and high NS group (P<0.01). In contrast, the percentage of express saccades for gaps of 250 ms did not differ significantly between the groups. There were no group differences on measures of accuracy or peak velocity.
All groups developed anticipatory saccades during the predictable protocol and there were no group differences on the percentage of anticipatory saccades. Accuracy of anticipatory saccades was also similar between the groups. There was a trend toward group differences in the peak velocities of anticipatory saccades; F(2,33)=3.18, P=0.056.
3.2. Errors on the oddball task
During the oddball task the percentage of errors (incorrect saccades executed toward the expected location, rather than the oddball target location) after the appearance of the oddball target demonstrated a subtle, but statistically significant difference between the groups (control: M=0.00% errors, SD=0.00; high NS: M=5.8%, SD=6.86; low NS: M=4.55%, SD=6.45; F(2,34)=4.17, P<0.05). Post hoc analyses revealed a significant difference between the control and high NS groups (P<0.05). There were no significant group differences for errors before the presentation of the oddball target.
3.3. Errors during the fixation, saccade-engagement and memory-guided tasks
Errors (i.e. inappropriate reflexive saccades toward the yellow cross) for each group and task are presented in Fig. 2. There was a significant main effect of Task F(3,96)=26.95, P<0.001 and a significant main effect of Group, F(1,32)=6.75, P<0.01. Post hoc analyses revealed that the high NS group (M=23.46%), made a significantly greater percentage of errors than the control group (M=7.18%: P<0.01).
Fig. 2.
Response suppression (% errors: inappropriate reflexive saccades to cues) for schizophrenia patients with high negative symptoms scores (high NS — filled); low negative symptoms scores (low NS — grey) and; and controls (CON — white) in the fixation; saccade-engagement; single memory-guided and two-step memory-guided. Error bars are standard errors.
3.4. Visually guided reflexive saccades and memory-guided saccades
3.4.1. Temporal measuresLatency and velocity means and standard deviations for each group are presented in Table 4. For all groups, latencies were significantly greater for memory-guided (M=426 ms) compared to reflexive (M=190 ms) saccades: F(1,29)=110.76 P<0.001. There were, however, no significant group differences or interactions.
Table 4. Mean values and standard deviations for latencies in seconds (s) and peak velocities (degrees/second) for reflexive and memory-guided saccades for each group
| Ocular motor paradigm | Control | High NS | Low NS |
|---|---|---|---|
| Mean±S.D. | Mean±S.D. | Mean±S.D. | |
| Reflexive saccades — 5° | |||
| Latency (s) | 0.20±0.02 | 0.19±0.02 | 0.19±0.02 |
| Peak velocity (degrees/s) | 229.61±43.92 | 261.24±70.66 | 218.12±27.22 |
| Reflexive saccades — 10° | |||
| Latency (s) | 0.20±0.02 | 0.17±0.02 | 0.18±0.02 |
| Peak velocity (degrees/s) | 264.84±63.63 | 403.59±76.11 | 349.14±57.74 |
| Reflexive saccades — 15° | |||
| Latency (s) | 0.20±0.03 | 0.19±0.02 | 0.19±0.03 |
| Peak velocity (degrees/s) | 435.97±83.13 | 446.08±71.77 | 411.89±73.28 |
| Memory-guided saccades — 5° | |||
| Latency (s) | 0.39±0.11 | 0.47±0.17 | 0.41±0.10 |
| Peak velocity (degrees/s) | 226.84±70.28 | 275.46±90.74 | 195.48±56.65 |
| Memory-guided saccades — 10° | |||
| Latency (s) | 0.43±0.18 | 0.50±0.25 | 0.45±0.12 |
| Peak velocity (degrees/s) | 307.50±99.66 | 349.48±75.66 | 267.53±62.74 |
| Memory-guided saccades — 15° | |||
| Latency (s) | 0.37±0.15 | 0.43±0.16 | 0.38±0.13 |
| Peak velocity (degrees/s) | 368.44±92.75 | 419.78±103.39 | 339.86±92.75 |
For all groups, peak velocities were significantly greater in the reflexive saccade task (M=349.24°/s) as compared to the memory-guided task (M=308.52°/s), F(1,30)=9.07 P<0.01. As expected, peak velocities significantly increased as saccade amplitude increased F(2,60)=217.67 P<0.001 and this increase was intensified in the reflexive task (5°: reflexive M=237.88°/s; memory-guided M=235.06°/s; 10°: reflexive M=375.05°/s; memory-guided M=310.88; 15°: reflexive M=434.78°/s; memory-guided M=379.61°/s: Task×Amplitude interaction, F(2,60)=13.81 P<0.001. There were no Group differences or Group interactions.
Means and standard deviations for primary saccade and final eye position gains, as well as intrasubject variability of gains (SD) are presented in Table 5 for each group for the reflexive saccade task. For reflexive saccade accuracy, primary saccade gain differed significantly for target amplitudes (5° M gain=0.96; 10° M gain=1.01; 15° M gain=0.96, F(2,60)=10.88, P<0.001), but there were no significant Group difference or interactions. Similarly, final eye position gain differed significantly for target amplitudes (5° M gain=0.97; 10° M gain=1.03; 15° M gain=1.02, F(2,62)=13.88, P<0.001), but there were no significant Group difference or interactions.
Table 5. Means and standard deviations for primary saccade and final eye position gains, as well as intrasubject variability of gains (SD) for reflexive and memory-guided saccades for each group
| Ocular motor paradigm | Control | High NS | Low NS |
|---|---|---|---|
| Mean±S.D. | Mean±S.D. | Mean±S.D. | |
| Reflexive saccades — 5° | |||
| Primary saccade gain | 0.95±0.09 | 1.01±0.21 | 0.96±0.09 |
| Final eye position gain | 0.95±0.11 | 0.96±0.16 | 1.0±0.08 |
| Variability of primary saccade gain (SD) | 0.17±0.05 | 0.19±0.09 | 0.22±0.07 |
| Variability of final eye position gain (SD) | 0.15±0.06 | 0.17±0.07 | 0.21±0.07 |
| Reflexive saccades — 10° | |||
| Primary saccade gain | 1.0±0.10 | 1.05±0.17 | 0.99±0.09 |
| Final eye position gain | 1.01±0.11 | 1.02±0.14 | 1.05±0.09 |
| Variability of primary saccade gain (SD) | 0.14±0.01 | 0.16±0.04 | 0.17±0.05 |
| Variability of final eye position gain (SD) | 0.13±0.04 | 0.12±0.06 | 0.15±0.06 |
| Reflexive saccades — 15° | |||
| Primary saccade gain | 0.97±0.10 | 0.98±0.16 | 0.93±0.10 |
| Final eye position gain | 1.01±0.12 | 1.0±0.14 | 1.05±0.11 |
| Variability of primary saccade gain (SD) | 0.10±0.04 | 0.18±0.03⁎⁎ | 0.14±0.03⁎ |
| Variability of final eye position gain (SD) | 0.08±0.04 | 0.12±0.04 | 0.09±0.04 |
Intrasubject variability of reflexive primary saccade gain differed significantly according to Amplitude [F(2,60)=18.76, P<0.001] and Group [F(1,30)=3.42,,P<0.05] and there was a Group by Amplitude interaction [F(4,60)=3.93, P<0.01]. Both patient groups demonstrated a trend toward increased variability, particularly for primary saccades of larger amplitudes (i.e. 15°). Intrasubject variability of final eye position gain differed significantly according to Amplitude [F(2,62)=40.37, P<0.001] and there was a Group by Amplitude interaction [F(4,62)=2.91, P<0.05].
Means and standard deviations for primary saccade and final eye position gains, as well as intrasubject variability of gains (SD) are presented in Table 6 for each group for the single and two-step memory-guided saccade tasks. For primacy saccade accuracy, there were no significant differences between single and two-step memory-guided saccades. There were no significant Group differences or Group interactions. For both groups, there was a significant difference between final eye position accuracy, in that saccades performed during the two-step memory-guided saccade task overshot the target (M gain=1.01, SD=0.15), as compared to saccades during the single memory-guided task, which undershot the target (M gain=0.96, SD=0.12), F(1,32)=7.594, P=0.01; however, there were no significant Group differences or Group interactions.
Table 6. Means and standard deviations for primary saccade and final eye position gains, as well as intrasubject variability of gains (SD) for single and two-step memory-guided saccades for each group
| Ocular motor paradigm | Control | High NS | Low NS |
|---|---|---|---|
| Mean±S.D. | Mean±S.D. | Mean±S.D. | |
| Single memory-guided saccades | |||
| Primary saccade gain | 0.97±0.11 | 0.90±0.09 | 0.86±0.18 |
| Final eye position gain | 0.99±0.12 | 0.94±0.07 | 0.92±0.13 |
| Variability of primary saccade gain (SD) | 0.28±0.15 | 0.26±0.13 | 0.21±0.08 |
| Variability of final eye position gain (SD) | 0.15±0.07 | 0.23±0.11 | 0.17±0.06 |
| Two-step memory-guided saccades | |||
| Primary saccade gain | 0.96±0.13 | 0.87±0.15 | 0.93±0.13 |
| Final eye position gain | 1.02±0.12 | 0.98±0.15 | 1.04±0.17 |
| Variability of primary saccade gain (SD) | 0.23±0.11 | 0.33±0.13 | 0.24±0.11 |
| Variability of final eye position gain (SD) | 0.17±0.06 | 0.24±0.10 | 0.17±0.04 |
There were no significant differences in the intrasubject variability of primary saccade gains between Groups or Tasks. Intrasubject variability of final eye positions differed significantly between groups (high NS, M=0.24; low NS, M=0.17; control, M=0.16: F(1,32)=4.48, P<05) and post hoc analyses revealed significantly increased variability in the high NS group, as compared to the control group (P<0.05) and low NS group (P=0.05).
3.5. Correlations with clinical measures
The only significant correlations between inhibitory errors and clinical ratings were positive correlations between the percentage of inhibitory errors during the fixation task and negative symptom ratings (according to PANSS negative symptom scores, r2=0.51, P<0.05 and SANS rating scores, r2=0.59, P<0.05). Other significant correlations were positive correlations between memory-guided saccade velocity and negative symptom rating scores (SANS) for saccades of 5° amplitudes (r2=0.45, P<0.05) and 10° amplitudes (r2=0.54, P<0.05). Total PANSS rating scores were also negatively correlated with primary saccade gain (r2=−0.61, P<0.01) and final eye position gain (r2=−0.53, P<0.05) for the two-step memory-guided saccades.
4. Discussion
This study sought to investigate inhibitory control and spatial working memory in patients with schizophrenia who have high and low negative symptoms (NS) scores. Greater inhibitory deficiency was observed in the high NS group in terms of increased errors during a saccadic oddball task and during a series of response suppression tasks. In addition, significant correlations were demonstrated between negative symptom scores and inhibitory errors during a fixation task.
By employing two sets of qualitatively different paradigms (fixation and saccade-engagement tasks versus single and two-step memory-guided tasks) we examined inhibitory control by exploring response suppression errors in schizophrenia patients with and without prominent negative symptoms. A previous study exploring the functional basis of inhibitory errors in a heterogeneous group of patients with schizophrenia reported that although schizophrenia patients were able to inhibit saccades to distractors during fixation, they had difficulty inhibiting inappropriate saccades during tasks that placed greater demands on working memory (e.g. the antisaccade and memory-guided saccade tasks) (Hutton et al., 2002). In contrast, we found patients in the high NS group made more errors than the control group during all tasks, including the fixation task. Moreover, inhibitory errors during the fixation tasks demonstrated a positive correlation with negative symptom scores, suggested that distractibility during fixation increases as negative symptom scores increase. This extends previous reports of increased inhibitory errors during the antisaccade task in those patients with more prominent negative symptoms (Tien et al., 1996, Nkam et al., 2001, Ettinger et al., 2004, Ettinger et al., 2006) and suggests an association between negative symptoms and inhibitory deficiencies.
Saccadic inhibition during tasks involving working memory (such as the antisaccade and memory-guided saccade tasks) is consistently related to dorsolateral prefrontal cortex dysfunction (Ploner et al., 1999, McDowell and Clementz, 2001). Distractibility during fixation may be suggestive of additional disturbances to the ‘fixation zone’ or rostral region of the superior colliculus, which acts to maintain fixation (in the face of distractions) by increasing the level of tonic inhibitory input to omnipause neurons in brainstem regions (Takahashi et al., 2005). Fixation neurons in the superior colliculus most likely receive inputs from ‘suppression regions’ of the frontal eye fields (Izawa et al., 2004a, Izawa et al., 2004b), as well as basal ganglia input from the substantia nigra pars reticulata, which is influenced by activity in the subthalamic nucleus during visual fixation (Hikosaka et al., 2000). Hence, it is possible that in addition to prefrontal dysfunction, any of these regions/pathways involved in saccade suppression during fixation may also be impaired in those patients with prominent negative symptoms.
During the oddball task, the high NS group, as compared to the control group, exhibited an increased number of inappropriate saccades to the no longer appropriate, but expected location, after the appearance of the oddball target. This suggests that patients have difficulty inhibiting a prepared saccade to an expected location, even after that saccade becomes inappropriate. Interestingly, similar impairments have been reported using this same saccadic oddball task in patients with Parkinson's disease (unpublished results). In the PD group, we interpreted these findings as response selection impairments that were manifestations of reduced inhibitory control. The neuropathological hallmark of PD is degeneration of nigrostriatal neurons in the basal ganglia. Many neurons in the basal ganglia (particularly within the caudate nucleus) become active before, not after, a particular ocular motor event (Apicella et al., 1992, Schultz et al., 1992) and it has been suggested that a major function of the basal ganglia may be to ‘open the gate’ and prepare target motor areas for an upcoming action based on expectation or prediction (Hikosaka et al., 2000). While healthy controls are capable of suppressing the prepared saccade to the expected (but no longer appropriate) location and generating a suitable saccade to the oddball target, the high NS group (like those patients with PD) appear to have difficulty suppressing prepared saccades. Hence, we can speculate that in the high NS group, the basal ganglia are actively preparing target motor areas for saccades, but the basal ganglia and dorsolateral prefrontal cortex pathways to the superior colliculus are failing to inhibit the triggering of the prepared saccade, despite changes in task demands (i.e. oddball target).
Both groups developed anticipatory behaviour at the same rate as controls. This finding is somewhat unexpected given the previous reports of increased anticipatory saccades during predictable paradigms in patients with schizophrenia (Karoumi et al., 1998, Spengler et al., 2006). Although not statistically significant, our pattern of findings suggesting a higher percentage of anticipatory saccades in the high NS group is consistent with previous reports (Karoumi et al., 1998). Both patient groups generated an increased percentage of express saccade during the gap task. In line with our findings, an increased occurrence of express saccades with gaps of 200 ms has previously been discussed in schizophrenia as a manifestation of deficient inhibitory control (Clementz, 1996). Our findings indicate that this effect is apparent for those patients with and without prominent negative symptoms.
Accuracy of memory-guided saccades was similar between the groups, even during the more complicated two-step memory-guided task (which involved an intervening visually-guided saccade during the delay period). Our correlation data demonstrated a significant inverse correlation between total PANSS scores and the accuracy of both the primary saccade and final eye position gain, suggesting that as schizophrenia symptoms increased in severity, the accuracy of two-step memory-guided saccades decreased. We did not find similar correlations in the single memory-guided task, perhaps indicating that in those patients with more severe schizophrenia symptoms, the neural circuitry involved in maintaining and updating spatial location information following intervening eye movements may be somewhat compromised, whether that be at the level of the parietal cortex, frontal eye fields or premotor cortex (see Karn et al., 1997).
In addition, the intrasubject variability of final eye position accuracy was increased for the high NS group, in both memory-guided conditions. Increased variability of memory-guided saccade accuracy has been previously reported in patients with lesions to the dorsolateral prefrontal cortex (Ploner et al., 1999); this finding may provide further evidence for the association between prefrontal dysfunction and negative symptoms.
There are two potential limitations to our findings. First, patients were not medication free. All patients were receiving antipsychotic medication and several patients were also receiving anticonvulsant medication (sodium valproate or carbamazepine). Of the few studies exploring the effects of atypical antipsychotics on eye movements, a reduction in inhibitory error rates during the antisaccade task (Burke and Reveley, 2002), as well as prolonged latency and decreased peak velocity and accuracy of visually-guided saccades (Sweeney et al., 1997) has been suggested. These effects are not entirely consistent with our findings. In addition, we were careful to only study patients receiving atypical antipsychotics and with no evidence of extrapyramidal side effects as high levels of dopaminergic blockade could produce effects similar to those seen in PD.
The second limitation was that patients with high negative symptom scores also had slightly higher scores on measures of positive symptoms and general psychopathology. Although we cannot ignore this as a potentially confounding factor, negative symptoms do not correlate with positive symptoms (Andreasen and Olsen, 1982, McGlashan and Fenton, 1992), suggesting they may involve independent processes. Our findings, particularly the significant correlations with negative symptom scores, do still support the notion that prominent negative symptoms are associated with distinct ocular motor abnormalities.
In conclusion, we have demonstrated increased inhibitory deficiencies in terms of increased response suppression errors, as well as increased response selection impairments, during the oddball task, in schizophrenia patients with prominent negative symptoms. The variability of accuracy was also increased in patients with prominent negative symptom scores. Collectively these findings provide further evidence to support the proposed association between prefrontal dysfunction and prominent negative symptoms.
References
- . Diagnostic and Statistical Manual of Mental Disorders, 4th ed., TR. Washington DC: APA; 2000;
- . Scale for the Assessment of Negative Symptoms (SANS). Iowa City, IA: University of Iowa; 1984;
- . Negative vs positive schizophrenia. Archives of General Psychiatry. 1982;39:789–794
- . Neuronal activity in monkey striatum related to the expectation of predictable environmental events. Journal of Neurophysiology. 1992;68:945–960
- . Differential relationships between positive and negative symptoms and neuropsychological deficits in schizophrenia. Schizophrenia Research. 1997;25:1–10
- . Parsing cognition in schizophrenia using saccadic eye movements: a selective overview. Neuropsychologia. 2001;39:742–756
- . Improved antisaccade performance with risperidone in schizophrenia. Journal of Neurology, Neurosurgery and Psychiatry. 2002;72:449–454
- . Scaling of movement velocity: a measure of neuromotor retardation in individuals with psychopathology. Psychophysiology. 1998;35:431–437
- . The ability to produce express saccades as a function of gap interval among schizophrenia patients. Experimental Brain Research. 1996;111:121–130
- . Psychophysiological measures of (dis)inhibition as liability indicators for schizophrenia. Psychophysiology. 1998;35:648–668
- . Volumetric neural correlates of antisaccade eye movements in first-episode psychosis. American Journal of Psychiatry. 2004;161:1918–1921
- . Antisaccade performance in monozygotic twins discordant for schizophrenia: the Maudsley twin study. American Journal of Psychiatry. 2006;163:543–545
- . Human express saccades: extremely short reaction times of goal directed eye movements. Experimental Brain Research. 1984;57:191–195
- . “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research. 1975;12:189–198
- . The prefrontal cortex and temporal integration. In: Jones EG, Peter A editor. Association and Auditory Cortices. Cerebral Cortex. vol. 4:New York: Plenum; 1985;p. 151–177
- . The Prefrontal Cortex: Anatomy, Physiology, and Neuropsychology of the Frontal Lobe. 3rd ed.. New York: Plenum; 1997;
- . Circuitry of primate prefrontal cortex and regulation of behaviour by representational memory. In: Mount-Castle VB editors. Handbook of Physiology, Section 1. vol. 5:Bethesda: American Physiological Society; 1987;p. 373–417
- . ECDUE Assessment Manual for Psychopharmacology. Rockville: National Institute of Mental Health; 1976;
- . The relationship between negative symptoms and neuropsychological performance. Biological Psychiatry. 1995;37:828–830
- . Role of the basal ganglia in the control of purposive saccadic eye movements. Physiology Review. 2000;80:953–978
- . Saccadic distractibility in first-episode schizophrenia. Neuropsychologia. 2002;40:1729–1736
- . Suppression of visually and memory-guided saccades induced by electrical stimulation of the monkey frontal eye field. I. Suppression of ipsilateral saccades. Journal of Neurophysiology. 2004;92:2248–2260
- . Suppression of visually and memory-guided saccades induced by electrical stimulation of the monkey frontal eye field. II. Suppression of bilateral saccades. Journal of Neurophysiology. 2004;92:2261–2273
- . Psychomotor slowing and planning deficits in schizophrenia. Schizophrenia Research. 2001;48:317–333
- . Reference frames in saccade targeting. Experimental Brain Research. 1997;115:267–282
- . Predictive saccade behaviour is enhanced in schizophrenia. Cognition. 1998;68:B81–B91
- . The Positive and Negative Syndrome Scale (PANSS) for schizophrenia. Schizophrenia Bulletin. 1987;13:261–275
- . Dementia Praecox and Paraphrenia. New York: Robert E Krieger Publishing Co. Inc.; 1919/1971;
- . The Neurology of Eye Movements. 3rd ed.. New York: Oxford University Press; 1999;
- . Fast reaction time predictive saccades among schizophrenia may be consistent with a failure of inhibition. Schizophrenia Research. 1997;24:243
- . Behavioral and brain imaging studies of saccadic performance in schizophrenia. Biological Psychiatry. 2001;57:5–22
- . Timing and amplitude of saccades during predictive saccadic tracking in schizophrenia. Psychophysiology. 1996;33:93–101
- . The positive–negative distinction in schizophrenia: a review of natural history validators. Archives of General Psychiatry. 1992;49:63–72
- . A new depression scale designed to be sensitive to change. British Journal of Psychiatry. 1979;134:382–389
- . The National Adult Reading Test (NART): Test Manual. Windsor: NFER-NELSON; 1982;
- . Saccadic and smooth-pursuit eye movements in deficit and non-deficit schizophrenia. Schizophrenia Research. 2001;48:145–153
- . Symptoms of the schizophrenic negative syndrome. British Journal of Psychiatry. 1996;169:209–212
- . Are there more than two syndromes in schizophrenia? A critique of the positive–negative dichotomy. British Journal of Psychiatry. 1992;161:335–343
- . Cortical control of saccades. Annals of Neurology. 1995;37:557–567
- . Errors of memory-guided saccades in humans with lesions of the frontal eye field and the dorsolateral prefrontal cortex. Journal of Neurophysiology. 1999;82:1086–1090
- . Cerebral perfusion correlates of negative symptomatology and parkinsonism in a sample of treatment-refractory schizophrenics: an exploratory 99m Tc-HMPAO SPET study. Schizophrenia Research. 1997;25:11–20
- . Volumetric measure of the frontal and temporal lobe regions in schizophrenia: relationship to negative symptoms. Archives of General Psychiatry. 2000;57:471–480
- . Neuronal activity in monkey ventral striatum related to the expectation of reward. Journal of Neuroscience. 1992;12:4595–4610
- . Drug induced extrapyramidal disorders: controlled studies of the relationship between behavioural change and the extrapyramisal symptoms produced by psychotropic drugs. Acta Psychiatrica Scandinavica. Supplementum. 1970;212:11–19
- . Evidence from increased anticipation of predictive saccades for a dysfunction of fronto-striatal circuits in obsessive–compulsive disorder. Psychiatry Research. 2006;143:77–88
- . Adverse effects of risperidone on eye movement activity: a comparison of risperidone and haloperidol in antipsychotic-naive schizophrenic patients. Neuropsychopharmacology. 1997;16:217–228
- . Synaptic inputs and their pathways from fixation and saccade zones of the superior colliculus to inhibitory burst neurons and pause neurons. Annals of the New York Academy of Sciences. 2005;1039:209–219
- . Eye movements and psychopathology in schizophrenia and bipolar disorder. Journal of Nervous and Mental Disease. 1996;184:331–338
- . Prefrontal cortex, negative symptoms, and schizophrenia: an MRI study. Psychiatry Research. Neuroimaging. 2001;108:65–78
- . Ocular motor differences between melancholic and non-melancholic depression. Journal of Affective Disorders. 2006;93:193–203
PII: S0165-1781(07)00034-0
doi:10.1016/j.psychres.2007.02.004
© 2007 Elsevier Ireland Ltd. All rights reserved.
