Motor automaticity in natural keyboard typing.

Certain features of everyday motor skills become automatic while others remain controlled. Here we use a novel keyboard typing task to investigate whether motor automaticity depends on the frequency of naturally learned motor sequences. Participants type five-letter strings that vary in their word and bigram (two-letter sequence) frequency in natural language, allowing us to examine the influence of prior exposure without laboratory training. Novel pseudo word strings are tested as well. We find greater sequence frequency in natural language is associated with faster inter-keypress intervals and lower temporal variability within the sequence. In contrast, latencies to initiate a sequence are slower for novel pseudo-word strings but are otherwise insensitive to natural word frequency. We also find individual differences in inter-keypress speed and variability are robust across frequency levels but are unrelated to conventional measures of typing skill. Our method establishes keyboard typing as a scalable, ethologically valid framework for probing features of a naturally acquired human motor skill. This research will help extend laboratory-based studies of motor sequence learning and sets the stage for future investigations of linguo-motor processes. Moreover, our findings demonstrate which features within naturally acquired motor sequences become automatic and that typing proficiency is not determined solely by automaticity.