tDCS Over the Motor Cortex Shows Differential Effects on Action and Object Words in Associative Word Learning in Healthy Aging

Abstract

Healthy aging is accompanied by a continuous decline in cognitive functions. For example, the ability to learn languages decreases with age, while the neurobiological underpinnings for the decline in learning abilities are not known exactly. Transcranial direct current stimulation (tDCS), in combination with appropriate experimental paradigms, is a well-established technique to investigate the mechanisms of learning. Based on previous results in young adults, we tested the suitability of an associative learning paradigm for the acquisition of action- and object-related words in a cohort of older participants. We applied tDCS to the motor cortex (MC) and hypothesized an involvement of the MC in learning action-related words. To test this, a cohort of 18 healthy, older participants (mean age 71) engaged in a computer-assisted associative word-learning paradigm, while tDCS stimulation (anodal, cathodal, sham) was applied to the left MC. Participants' task performance was quantified in a randomized, cross-over experimental design. Participants successfully learned novel words, correctly translating 39.22% of the words after 1 h of training under sham stimulation. Task performance correlated with scores for declarative verbal learning and logical reasoning. Overall, tDCS did not influence associative word learning, but a specific influence was observed of cathodal tDCS on learning of action-related words during the NMDA-dependent stimulation period. Successful learning of a novel lexicon with associative learning in older participants can only be achieved when the learning procedure is changed in several aspects, relative to young subjects. Learning success showed large inter-individual variance which was dependent on non-linguistic as well as linguistic cognitive functions. Intriguingly, cathodal tDCS influenced the acquisition of action-related words in the NMDA-dependent stimulation period. However, the effect was not specific for the associative learning principle, suggesting more neurobiological fragility of learning in healthy aging compared with young persons.

Keywords: associative word learning; healthy aging; language functions; motor cortex; transcranial direct current stimulation.

Figures

Figure 1

Word learning paradigm. (A) Participants were shown a photographic illustration depicting an action/object paired with a spoken pseudoword. They had to decide intuitively if the presented coupling was a correct or incorrect match. Two-hundred milliseconds after onset of the sound, the picture appeared. Responses had to be given within a time window of 3000 ms. The interstimulus interval (ISI) was 2000 ms between two trials. (B) One learning session consisted of 680 single trials, subdivided into five blocks with a 2 min break. Each pseudoword was shown 10 times with the “correct” action/object (depicted by two different images) as well as 10 times with different “incorrect” images. (C) Study design: prior to the first learning session participants were screened with a neuropsychological test battery (NP) and were introduced to the learning paradigm with a small lexical pre-test (PT) consisting of five words. Each participant completed three learning sessions on three different days with three different lexicon sets. During the task, participants received anodal, cathodal or sham tDCS in a randomized order across learning sessions (1 mA for 20 min; double blind application). At the start of each session the stimulation side over the left motor cortex (MC) was localized using transcranial magnetic stimulation (TMS). tDCS stimulation and the learning paradigm started simultaneously, the latter exceeding the stimulation for approximately 20 min. After each session, patients were asked to translate the acquired pseudowords into notions of their native language (T). Abbreviations: NP, neuropsychological testing; PT, pre-test; T, translation test.

Figure 2

Changes in accuracy depending on stimulation type and word type. Mean accuracy scores over the course of five learning blocks (B1–5) split by type of tDCS stimulation (A = anodal = triangles; C = cathodal = rectangles; Sham = circles). Dashed light gray lines depict action words, dotted dark gray lines depict object words; the bold black line with circles depicts overall mean accuracy.

Figure 3

Cognitive task performance and word learning. Left side: correlations between learning success measured as mean percentage of correct responses in block 5 and the performance scores of (A) VLMT (verbal learning and memory test) respectively (B) logical reasoning. Right side: correlations between number of correctly translated words and performance scores of (C) VLMT or (D) logical reasoning.