This paper examines possible influences of language on thought in the domain of spatial reasoning. Language communities differ in their stock of reference frames (coordinate systems to reference locations and directions). English typically uses egocentrically-defined axes (“leftright”). Other languages like Tseltal lack such a system but use geocentrically-defined axes ("north-south"). We ask whether the lexical resources available in one’s language determine the availability or salience of certain spatial concepts. Does a “left-right ” lexical gap translate to a conceptual gap? In three experiments, we compared Tseltal speakers ’ ability to solve spatial problems requiring an egocentric frame of reference to ones requiring a geocentric frame of reference. We found that Tseltal speakers were above chance in solving the egocentric problems, demonstrating that a lexical gap does not necessarily lead to a conceptual gap. Furthermore, participants were statistically better on the egocentric version in two of the three experiments. These results speak against some current versions of linguistic relativity.

Controversy over the nature of the innate endowment for language is an ongoing theme in the literature on language acquisition. While it might appear that this controversy reflects the existence of fundamental unresolved issues in the field, the areas of disagreement are more restricted than initial appearances suggest.

Human infants show a preference for listening to speech, but little is known about how infants listen to other naturally occurring sounds. Here, we test infants ’ listening bias for speech against a range of naturally occurring sounds that share properties of speech to varying extents and we aim to better characterize the speech properties that attract infant attention. We compared 3-month-olds’ listening patterns for five types of sounds: nonnative speech, rhesus macaque vocalizations, human noncommunicative vocalizations, human communicative nonspeech vocalizations, and environmental sounds. Across three experiments, 3-month-olds preferred speech to the other four types of sounds. The set of acoustic properties we measured—pitch, peak amplitude, nonzero-root mean square amplitude, frequency difference and amplitude variance—did not predict infant looking time. Our results demonstrate that young infants attend selectively to speech over many other naturally occurring stimuli, an important tool for learning language. For human infants to learn language, they must distinguish the sounds (and gestures) that are relevant for language acquisition. Infants as young as newborns possess listening biases that direct them towards human speech. One line of research shows that infants listen longer to speech compared with a range of synthetic sounds such as filtered speech (Spence & DeCasper, 1987),

Reprints and permission: