NDA@SZTAKI

Differential vulnerability of different forms of skill learning in Parkinson?s disease

Lukács, Ágnes

Kemény, Ferenc

Demeter, Gyula

Valálik, István

Racsmány, Mihály

2014

BF Psychology / lélektan

BF13 Memory and learning / emlékezet, tanulás

RC0321 Neuroscience. Biological psychiatry. Neuropsychiatry / idegkórtan, neurológia, pszichiátria

RD Surgery / sebészet

The striatal dopaminergic dysfunction in Parkinson?s disease (PD) has been
associated with deficits in skill learning in a number of studies, but the results are inconclusive
so far. Motor sequence learning (especially sequence-specific learning) is found to be deficient in
the majority of studies using the Serial Reaction Time Task (SRTT; Siegert, Taylor, Weatherall,
& Abernethy, 2006; Jackson et al., 1995; Ferraro, Balota and Connor, 1993; Pascual-Leone et
al., 1993, Muslimovic et al., 2007; Gobel et al., 2013; but see Kwak et al., 2012), although
results are contradictory when verbal response is required instead of button presses (Westwater
et al. 1998; Smith, Siegert and McDowall 2001). While problems with motor sequences seem to
be prevalent, PD patients show intact performance on Artificial Grammar Learning (AGL) tasks,
suggesting that the sequencing problem may be response type- or task type-dependent (Smith,
Siegert and McDowall 2001; Witt, Nühsman and Deuschl, 2002) Acquisition of nonsequential
probabilistic associations also seems to be vulnerable as evidenced by impaired PD performance
on a probabilistic category learning task (Knowlton, Mangels et al., 1996; Shohamy, Myers,
Onlaor, & Gluck, 2004). Our aim was to explore the nature of the skill learning deficit by testing
different types of skill learning (sequential versus nonsequential, motor versus verbal) in the
same group of Parkinson?s patients. 14 patients with PD (mean age: 59.77 range: 45.5-74) were
compared to age-matched typical adults using 1) a Serial Reaction Time Task (SRTT) testing the
learning of motor sequences, 2) an Artificial Grammar Learning (AGL) task testing the
extraction of regularities from auditory sequences and 3) a Weather prediction task (PCL-WP),
testing probabilistic category learning in a non-sequential task. In motor sequence learning on the
SRTT task, the two groups did not differ in accuracy; PD patients were generally slower, and
analysis of z-transformed reaction times showed no evidence of sequence learning in PD. A
deficit in artificial grammar learning was present only as a tendency in the PD group. The PD
group showed evidence of learning on the PCL task, and their learning performance was not
statistically different from that of the control group. These results partly support and also extend
previous findings suggesting that motor skill learning is vulnerable in PD, while other forms of
skill learning are less prone to impairment. Results are also in line with previous assumptions
that mechanisms underlying artificial grammar learning and probabilistic categorization do not
depend on the striatum (Reber & Squire, 1999; Skosnik et al., 2002).

Conference or Workshop Item

PeerReviewed

text

http://real.mtak.hu/26637/1/SFN_2014_abstract.pdf

Lukács, Ágnes and Kemény, Ferenc and Demeter, Gyula and Valálik, István and Racsmány, Mihály (2014) Differential vulnerability of different forms of skill learning in Parkinson?s disease. In: Annual Meeting of the Society for Neuroscience, 15-19 November, Washington, DC.

http://real.mtak.hu/26637/