Effects of Object Size on Unimanual and Bimanual Movements in Patients With Schizophrenia

Research Design

A counterbalanced repeated-measures design was used. Participants were randomly assigned to either the large-object condition or the small-object condition first and the other condition second. The study protocol was approved by the Human Research Ethics Committee of Jianan Mental Hospital in Taiwan. All participants signed informed consents.

Participants

We used convenience sampling to recruit patients with mild schizophrenia and healthy control participants. The two groups were matched on age and gender. Patients were recruited from the day care or chronic wards of a psychiatric teaching hospital. Healthy control participants were recruited from a university. The inclusion criteria for participants with mild schizophrenia were (1) diagnosis of schizophrenia or schizoaffective disorder by a psychiatrist on the basis of criteria of the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (American Psychiatric Association, 1994); (2) absence of extrapyramidal motor symptoms according to results on the Extrapyramidal Symptom Rating Scale (Chouinard & Margolese, 2005; Chouinard, Ross-Chouinard, Annable, & Jones, 1980); (3) stable psychiatric symptoms—that is, the use of the same oral or injected antipsychotics for at least the 4 wk immediately before the study began; (4) sufficient cognitive ability—that is, a score >24 on the Mini-Mental State Examination (Folstein, Folstein, & McHugh, 1975; Guo et al., 1988); (5) right-handedness; (6) no substance abuse; and (7) no neurological or musculoskeletal diseases, diabetes mellitus, or other diseases that might affect manual movements. We used the Positive and Negative Symptom Scale (Cheng, Ho, Chang, Lan, & Hwu, 1996; Kay, Fiszbein, & Opler, 1987) to evaluate symptom severity in patients with schizophrenia; higher scores mean more severe symptoms. The inclusion criteria for healthy control participants were Criteria 4 through 7.

Task Setup and Kinematic Measures

All task materials were placed on a height-adjustable table. A stationary pin (target) mounted on a base was centered 60.0 cm from the body midline of the participant (Figure 1). The tip of the pin was 27.0 cm above the table. Two cylindrical objects were located close to the participant’s midline near the edge of the table. The right object was 3.0 cm wide. Two sizes of the left object were used: large (7.0 cm wide) and small (3.0 cm wide). We manipulated the size of the left object because the literature (Riek, Tresilian, Mon-Williams, Coppard, & Carson, 2003; Srinivasan & Martin, 2010) suggests that a right-handed person tends to look at the left object first. The large and small objects had the same height (1.4 cm) and similar weights (21.5 g and 20.9 g). Inner diameters (4.7 cm and 2.0 cm) were two-thirds of the outer diameters (7.0 cm and 3.0 cm) for each object. For the starting position, the participant was required to place the right hand 40.0 cm away from the right object and to place the left hand 20.0 cm away from the left object. Unequal aiming distances for each hand were set to simulate an actual work situation.

OPEN IN VIEWER

Figure 1.

Experimental setup for the large-object (7.0 cm) condition of the left hand showing (A) starting position and (B) assembly movement.

Movements were recorded by an optical motion capture system (Qualisys AB, Gothenburg, Sweden) containing six cameras and a stationary measurement computer. The cameras use light-emitting diodes to generate invisible infrared flashes and illuminate reflective markers, after which they capture three-dimensional movements of the markers. Movements were recorded at 70 Hz. Six markers (0.4 cm in diameter; three on each hand) were attached to the participant to capture hand movements: on the dorsal side of the thumbnail and index fingernail and on the metacarpal base of the long finger (wrist). Three additional markers were attached to the upper side of the two objects to capture the position of the object planes. One marker was attached to the top of the target pin to capture its position. Motion capture data in the computer were tracked by Qualisys Track Manager software (Qualisys AB, Gothenburg, Sweden) and then streamed to Matlab 7.1 (Mathworks, Natick, MA) for kinematic calculation.

Procedures

Each participant sat in front of a table with the table height adjusted close to his or her elbow. The participant put each hand on the corresponding starting position. Upon receiving the starting signal, the participant moved both hands inward to grasp the objects (prehension) and then moved both hands forward to place objects on the target pin (assembly). The participant was asked to grasp the object with the thumb and index finger and to place the left object first and then the right one. In addition, the participant was required to perform the task as quickly and accurately as possible. After a short practice session, the participant did five trials in each condition with a 5-min break between conditions.

Variable Definitions

We dissected the movement sequence into prehension and assembly and identified four critical events (time points of movement onset and movement end). The definitions of critical events were (1) prehension onset, which was when the initial wrist velocity reached a speed that was 5% of the prehensile peak velocity; (2) prehension end, which was when the thumb velocity dropped to 0 mm/s and the average position of the three markers on the object did not change; (3) assembly onset, which was when the wrist velocity reached a speed that was 5% of the assembly peak velocity; and (4) assembly end (goal achievement), which was when the left object dropped to the pin for the left hand and when the right object dropped to the pin for the right hand.

To characterize unimanual movements, we calculated movement time and peak velocity to represent movement speed and forcefulness. Movement time is defined as the time interval between movement onset and movement end. A shorter movement time means a faster movement. Peak velocity is the maximal instantaneous velocity during a movement. A higher peak velocity means greater force generation (Lin, Wu, Lin, & Chang, 2008).

For bimanual coordination, we calculated the timing interval between both hands at critical events, including prehension onset, prehension end, assembly onset, and assembly end. A shorter interval indicates more synchronous bimanual movements and thus better bimanual coordination (Wu et al., 2009).

Data Analysis

We used two-way analysis of variance (ANOVA) with one between factor (Group: patients with mild schizophrenia vs. healthy control participants) and one within factor (Size: large vs. small). Significance (two-tailed) was set at .05. Effect size, η², was calculated to estimate the magnitude of the effects. Values for small, medium, and large effect sizes are η² = .01, .06, and .14, respectively (Cohen, 1988).

Participant Characteristics

Fifteen patients with mild schizophrenia and 15 healthy control participants were enrolled in our study. No participants withdrew. Table 1 shows demographic and clinical characteristics of participants. The age- and gender-matched control participants had significantly higher education levels and scores on the Mini-Mental State Examination than the patients.

Group Differences

Two-way ANOVA showed no significant Group × Size interaction effects for any variables (Table 2). For unimanual movements, significant group effects were found for prehensile and assembly movement times and for assembly peak velocities for both hands. For bimanual movements, a significant group effect was found for the timing interval at assembly end. Compared with the control participants, the patients with schizophrenia had longer prehensile and assembly movement times, lower assembly peak velocities for both hands, and a longer timing interval between both hands at assembly end.

Effects of Object Size

For unimanual movements, significant size effects were found for prehensile and assembly movement times, for the assembly peak velocities for both hands, and for the prehensile peak velocity for the right hand (Table 2). For bimanual movements, significant size effects were found for the timing intervals at prehension end and assembly end. Compared with the small-object condition, the large-object condition yielded shorter prehensile and assembly movement times, higher assembly peak velocities for both hands, a higher prehensile peak velocity for the right hand, and shorter timing intervals between both hands at prehension end and assembly end.

Patients’ Movements in the Bimanual Assembly Task

In the bimanual assembly task, the patients had slower and less forceful unimanual movements, as well as less coordinated bimanual movements at assembly end, than the control participants. These results are consistent with those of other studies that reported movement disorders in patients with schizophrenia (Delevoye-Turrell et al., 2003; Haddad & Mattay, 2011; Henkel et al., 2004; Koning et al., 2010; MacRae, 2005; Pappa & Dazzan, 2009). Our study extends earlier reports by providing information on how the patients generated bimanual functional movements in terms of speed, forcefulness, and bimanual coordination. We quantified the patients’ movement process instead of merely showing movement outcomes. The results of movement impairments may be explained by patients’ basal ganglia dysfunction (Shenton, Dickey, Frumin, & McCarley, 2001; Torrey, 2002), which disables the internal modulation of motor programs from the cerebral cortex (Canning, 2010) and thus leads to impaired timing control over unimanual and bimanual movements (Debaere, Wenderoth, Sunaert, Van Hecke, & Swinnen, 2003; Majsak, Kaminski, Gentile, & Flanagan, 1998).

We used the Extrapyramidal Symptom Rating Scale to ensure that we included only patients with schizophrenia who had no extrapyramidal motor symptoms. The results of our kinematic analysis showed that, compared with control participants, the patients without extrapyramidal motor symptoms still had slower, less forceful, and less coordinated movements. Our findings are consistent with previous studies that indicated that instrumental measures are more sensitive to subclinical motor impairments than observer-based ratings (Caligiuri & Lohr, 1994; Cortese et al., 2005; Pappa & Dazzan, 2009). Our findings illustrate the importance of kinematic measures in identifying patients’ movement impairments and detecting their movement responses to therapy. We suggest that occupational therapy practitioners consider movement rehabilitation for patients with mild schizophrenia.

Effects of Object Size

We manipulated the size of the left object and found that movements of both hands were affected. A large object induced faster and more forceful unimanual movements, as well as more coordinated bimanual movements, than did a small object in patients with mild schizophrenia and in healthy control participants. The result that patients responded to the object property in a manner similar to the healthy control participants is in accordance with earlier research (Delevoye-Turrell et al., 2003).

Moreover, the effect of object size is consistent with previous studies (Adam, 1992; Bingham et al., 2008; Fitts, 1954; Fowler et al., 1991; Kelso et al., 1979). We extend earlier findings by showing that this effect also exists in patients with schizophrenia. Rose and Christina (2006) suggested that when people reach for objects of various sizes, the visual system provides the brain with information to correct endpoint errors of the movement trajectory and thus to ensure that the movement is accurate. More time to correct movements is needed when objects are smaller (Shumway-Cook & Woollacott, 2007), which results in a longer movement time. For bimanual movements, because both upper limbs rely on visual information to correct movements, decreasing the object size makes it more difficult for the visual system to simultaneously guide both hands (Bingham et al., 2008). People need to fixate on one object first to correct endpoint errors of the movement trajectory and then fixate on the other object for the same purpose (Bingham et al., 2008; Riek et al., 2003; Srinivasan & Martin, 2010), thus leading to decreased bimanual synchronization at movement ends.

We found that increasing the object size induced higher peak velocities in assembly movements. A few studies on the effect of object size on peak velocities reported contradictory results (Adam, 1992; Bingham et al., 2008; Jackson et al., 1999). One possible reason is that different studies had different movement requirements. Our study and others (Adam, 1992; Jackson et al., 1999) that used pointing or prehensile movements reported that increasing the object size elicited higher peak velocities. However, Bingham et al. (2008), who not only required prehensile movement but also required fingers to contact designated parts of the object, found no effects of object size on peak velocities. We hypothesize that complex movements weaken the impact of object size on peak velocities; this hypothesis needs additional examination.