Effect of meaning on apraxic finger imitation deficits
Introduction
Apraxia is a disorder of motor cognition which most often occurs after left hemisphere (LH) stroke (Donkervoort et al., 2000), but has also been reported after right hemisphere (RH) stroke (Donkervoort et al., 2000, Goldmann Gross and Grossman, 2008).
Apraxia is characterized by a bilateral impairment of purposeful, skilled movements, including imitation. Importantly, as a higher motor deficit apraxia cannot be fully accounted for by primary deficits of the sensorimotor system or disturbed communication (Dovern et al., 2012). Tests of gesture imitation are frequently used for the (bed-side) assessment of apraxia (Goldenberg, 2008). However, it remains a challenge to differentiate whether an observed imitation deficit is due to apraxia or results from (often co-morbid) aphasia and associated semantic deficits when meaningful (MF) gestures have to be imitated (Goldenberg, 2008). Accordingly, for diagnosing apraxic imitation deficits, tests which employ meaningless (ML) gestures are recommended, since ML gestures are supposed to be unfamiliar/novel and hence should not be represented in long-term memory (Goldenberg, 1996, Mengotti et al., 2013). Current cognitive models of gesture imitation suggest that ML (unfamiliar/novel) gestures need to be processed via a direct (non-semantic) route without access to action semantics, while MF (familiar) gestures can also be reproduced via an indirect (semantic) route accessing pre-existing motor representations and action semantics (Cubelli et al., 2000, Rothi et al., 1991, Rumiati and Tessari, 2002). Note that the direct route is termed “direct” as it bypasses semantic processing. It should be stressed that this does not exclude other cognitive processes like deployment of attention, visuo-spatial transformations, or body part coding (Goldenberg, 2008).
Two imitation tests devised by Goldenberg are commonly used to assess gesture imitation in stroke patients (Goldenberg, 1996): the imitation of hand positions and the imitation of finger configurations. The gestures used in both tests are supposed to be ML, i.e., unfamiliar/novel gestures. While this is not questioned for the test of imitating hand positions, it is unclear whether all finger configurations used in the finger imitation test are truly ML. For example, finger configurations F01, F02, F06 and F08 (see Fig. 1) correspond to counting gestures that are known to activate number knowledge in healthy adults (for review, see (Fischer and Brugger, 2011). To address this important concern, we here investigated the effect of meaning on finger imitation deficits in a large cohort of stroke patients (n=368), thereby probing current cognitive models of gesture imitation which were based on much smaller patient samples (e.g., Cubelli et al., 2000; Mengotti et al., 2013; Tessari et al., 2007). Furthermore, the current investigation sheds light on the finger imitation test, which has a special standing within the diagnostics of limb apraxia, since-in contrast to most other commonly used limb apraxia tests (e.g., hand imitation)-the finger imitation test seems to be sensitive not only to imitation deficits after damage to the motor-dominant left hemisphere, but also to deficits resulting from right hemisphere damage (Goldenberg et al., 2009, Goldenberg and Strauss, 2002).
To this end, we first assessed whether healthy subjects (n=45) classified the 10 finger gestures of the Goldenberg finger imitation test (Goldenberg, 1996) as ML or MF. We next evaluated, retrospectively, in 255 patients with LH stroke and 113 patients with RH stroke whether their imitation performance of these finger gestures was influenced by meaning. We found that only two of the 10 gestures were perceived as clearly ML and that meaningfulness improved imitation scores of apraxic patients. Furthermore, the severity of aphasic deficits in LH stroke patients significantly influenced their imitation of MF gestures. Our results have implications for the diagnosis of imitation deficits and for theoretical models of gesture imitation.
Section snippets
The test of imitating finger configurations
In the test of imitating finger configurations by Goldenberg (1996), the examiner sits opposite to the patient and demonstrates each finger gesture (in the order F01–F10, see Fig. 1) with the hand opposite to the patient's non-paretic ipsilesional hand, which the patient is supposed to use for imitation. After the first demonstration of each finger gesture, the examiner forms a fist (neutral gesture) and the patient is asked to imitate the previously shown finger gesture. Two points are
Meaning affects imitation of finger configurations especially in patients with imitation deficits
In addition to the expected main effect of IMITATION DEFICIT (F (1, 364)=512.68, p<0.001; note that this factor – by definition – separated stroke patients with and without finger imitation deficits), the ANOVA revealed a significant main effect of MEANING (F (1, 364)=157.05, p<0.001): overall, the two ML finger gestures were imitated worse than the three MF finger gestures (mean imitation score±S.E.M: ML finger gestures: 1.69±0.03; MF finger gestures: 1.86±0.02). The interaction MEANING by
Discussion
Our main finding in a large sample of stroke patients that meaning of a gesture affects finger imitation corroborates current models of gesture imitation. Stroke patients’ performance in the test of imitating finger configurations by Goldenberg (1996) was affected by the degree of meaning that healthy subjects associate with these finger configurations. Of the 10 finger gestures applied in the finger imitation test, two finger gestures were clearly judged ML, while three finger gestures were
Limitations
Despite the large sample of stroke patients investigated (n=368) and the fact that our main finding (the influence of meaning on imitation performance especially in stroke patients with a finger imitation deficit) was confirmed by three different types of analyses, some limitations of our study need to be considered.
The gestures F03 and F09 were the two gestures that were clearly classified as ML. One could argue that these two gestures are also the only gestures demanding a distinction between
Conclusion
Our findings in a large cohort of stroke patients support current cognitive models of imitation (Cubelli et al., 2000, Rothi et al., 1991, Rumiati and Tessari, 2002). Furthermore, our data confirm the importance of including the imitation of ML gestures into the apraxia assessment of stroke patients because ML gesture imitation is not influenced by the severity of aphasic deficits in LH stroke patients and is particularly sensitive to detect imitation deficits (independent of the hemisphere
Conflict of interest
The authors declare no competing financial interests.
Acknowledgements
We thank our healthy participants and patients, the examiners and our colleagues at the Neurology Department of the University of Cologne and the Institute of Neuroscience at the Research Centre Jülich. GRF gratefully acknowledges support from the Marga and Walter Boll Foundation. MHF is supported by the German Research Council (DFG) Grant FI 1915/2-1 “Manumerical cognition”.
References (37)
- et al.
Crossed aphasia and related anomalies of cerebral organization: case reports and a genetic hypothesis
Brain Lang.
(1996) - et al.
Double dissociation between meaningful and meaningless gesture reproduction in apraxia
Cortex
(2001) - et al.
List effect in apraxia assessment
Neurosci. Lett.
(2006) - et al.
Cognition in action: testing a model of limb apraxia
Brain Cogn.
(2000) Apraxia
- et al.
The meaning of meaningless gestures: a study of visuo-imitative apraxia
Neuropsychologia
(1997) - et al.
Severity of neglect predicts accuracy of imitation in patients with right hemisphere lesions
Neuropsychologia
(2009) - et al.
A cross-validation study on the token test
Neuropsychologia
(1973) - et al.
Clinical validation of a new test for aphasia: an experimental study on the token test
Cortex
(1966) - et al.
Token test scores of three matched patient groups: left brain-damaged with Aphasia; right brain-damaged without Aphasia; non-brain-damaged
Cortex
(1969)