Use this URL to cite or link to this record in EThOS:
Title: The influence of language on spatial memory and visual attention
Author: Gudde, Harmen
ISNI:       0000 0004 7227 4502
Awarding Body: University of East Anglia
Current Institution: University of East Anglia
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Access from Institution:
This thesis examines the relationship between language and non-linguistic processes. The experimental work presented, focusses on the influence of language on two non-linguistic processes: spatial memory and visual attention. In the first series of experiments, the influence of spatial demonstratives (this/that) and possessives (my/your) on memory for object location was examined in four experiments, using an adapted version of the memory game procedure (Coventry et al., 2008, 2014). The experiments were designed to test between different models regarding how language affects memory: the Expectation model, the Congruence model, and the Attention-allocation model. Over a series of experiments, our data supports the Expectation model, which suggests, consistent with models of predictive coding (cf., Lupyan & Clark, 2015), that memory for object location is a concatenation of the actual location and the expected location. The expectation of a location can be elicited by language use (e.g., demonstrative or possessive pronouns). The second series of experiments examined demonstratives and memory in English and Japanese. We chose Japanese, because it purportedly employs a three-demonstrative system, compared to a binary system as in English (this, that). Three-way systems can be used to explicitly encode parameters that are not encoded in English, for example the position of a conspecific. In four experiments, we wanted to test whether a system as different as the Japanese demonstrative system is from English, has a similar influence on non-linguistic cognition. To this aim, we had to first experimentally establish which parameters are encoded in the Japanese demonstrative system. Second, we tested how this three-term demonstrative system acted in light of the Expectation model. The idea that Japanese demonstratives encode the position of a conspecific, which we confirmed in this study, poses an interesting problem for the Expectation model. The Expectation model works via the idea of an expected location; but the expected location calculated from a speaker gives a contradicting expectation value to the expected location from a hearer. Our memory data did not completely support any of the current models. However, interestingly, the position effect found in Japanese was also apparent in English. This might suggest that demonstrative pronoun systems, despite the fact that they seem different, could be based on universal mechanisms. However, the effects we found were stronger in Japanese, suggesting the weight of a parameter (such as position) might be influenced by whether or not a language explicitly codes the parameter. In the last experiment, we considered the influence of language on visual attention. Specifically, we examined if language expressing different spatial frames of reference affect how people look at visual scenes. The results showed different eye-movement patterns for different frames of reference (i.e., intrinsic vs. relative). These eye-movement signatures were consistent with participants’ verbal descriptions and persisted throughout the trials. We show for the first time that different reference frames, expressed in language, elicit distinguishable eye-movement patterns. The work presented in this thesis shows effects of language on memory for object location and visual attention. Effects of language on memory for object location were consistent with models of predictive coding. Furthermore, despite the fact that English and Japanese employ different demonstrative systems, results for both languages were remarkably similar. These results could indicate universal parameters underlying demonstrative systems, but perhaps parameters differentially weighted, as a function of whether or not they are explicitly encoded in a language. Finally, we showed that spatial language (prepositions) guide visual attention. To our knowledge this is the first time frames of reference are associated with identifyable eye-movement patterns. The results are discussed and situated in current literature, with theoretical implications and directions for future research highlighted.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available