Eye tracking studies of reference in real-world contexts demonstrate just how "eager" the language comprehension system is. As soon as words are segmented from the speech stream, hearers try to map them onto possible referents in the world. When a referring phrase is long and complex (for example: "The tall red cup with the gold trim") there may be quite a lot of distance between the time the hearer begins the mapping process, and the point at which there is enough information in the linguistic stream to distinguish the object from other similar objects.

This pair of graphs shows the pattern of eye movements to four different objects in the visual display as a hearer responds to the instruction "Pick up the tall glass." The graphs show how the patterns change, depending on which objects are in the display. Eye gaze to the referent (the tall glass) is shown in black. (From Sedivy et al., 1999).

In such cases of temporary "referential indeterminacy", hearers try to make educated guesses about the intended referent. What information do they use to do this? Work in our lab has shown that they take into consideration, among other things, the other objects in the visual array, expectations about the amount of information the speaker is likely to provide about each of the objects, and even considerations of the likely linguistic capabilities of the speaker.

Grodner, D. & Sedivy, J. (In press). The effects of speaker-specific information on pragmatic inferences. In N. Pearlmutter & E. Gibson (eds). The Processing and Acquisition of Reference. MIT Press: Cambridge, MA.

Sedivy, J. (2003). Pragmatic versus form-based accounts of referential contrast: Evidence for effects of informativity expectations. Journal of Psycholinguistic Research.

Sedivy, J, Tanenhaus, M., Chambers, C., & Carlson, G. (1999). Achieving incremental semantic interpretation through contextual representation. Cognition, 71, 109-147.

Adult hearers are good at using their understanding of how speakers behave in referential situations to help them in processing the incoming linguistic stream. What about children? They must cope with immature processing systems, a less sophisticated understanding of their conversational partner, and often, incomplete knowledge of what the words mean. There are many unanswered questions about how children's development proceeds. For example, do young children have enough understanding about how communication works to help them make educated guesses about the meanings of new words? Are they able to use their knowledge of a speaker's likely behavior to help them process language, or do they interpret language in an "egocentric" way. Some of the work in our lab has begun to address questions like these.

Nadig, A. & Sedivy, J. (2002). Evidence of perspective-taking constraints in children's on-line reference resolution. Psychological Science, 13 (4), pp. 329-336.

Nadig. A., Sedivy, J., Joshi, A. & Bortfeld, H. (2003). The development of discourse constraints on the interpretation of adjectives. Proceedings of the 27th BU Conference on Language Development. Boston: Cascadilla Press.

The ease with which we recognize most words we know when they are spoken belies the intricate mechanism by which we do so. We must match information from the incoming speech stream against a memory store containing tens of thousands of words. Complicating this process is the fact that our memory representations of these words form a tightly interconnected web such that, along with pulling out the word that "matches" the acoustic stream, we also pull out words that are related in sound and meaning. Some of the work in our lab has focused on how this delicate process works.

In the display that you see here, when hearers respond to "Click on the hammock", they show fleeting eye movements to the nail, revealing the connections in memory among words with similar sound and meaning ("hammock" activates "hammer" through sound similarity, which then activates "nail" through its meaning relationship). (From Yee & Sedivy, 2006).

Yee, E., Blumstein, S. & Sedivy, J. (In press). Lexical activation in Broca's and Wernicke's Aphasia: Evidence from eye movements. Journal of Cognitive Neuroscience.

Yee, E. & Sedivy, J. (2006). Eye movements reveal transient semantic activation during spoken word recognition. Journal of Experimental Psychology: Learning, Memory and Cognition, 32, 1-14.

Myung, J., Blumstein, S. & Sedivy, J. (2006). Playing on the Typewriter, Typing on the Piano: Manipulation Knowledge of Objects. Cognition 98, 224-243

In referring to things and people, we use a variety of different linguistic expressions, including pronouns such as "he" or "she". When there are numerous people that are being talked about in the discourse, how do we know who the pronoun refers to? It turns out that hearers normally expect pronouns to be referring back to things or people that are foregrounded, or highly salient in the discourse. Some of the eyetracking work in our lab uses spoken "stories" accompanied by illustrations top explore linguistic factors that make discourse referents more or less salient, and what happens with visual attention when a pronoun is used.

Heller, D., Yee, E. & Sedivy, J. (2007) What do eye movements tell us about reference resolution? Using ambiguous pronouns to evaluate linking hypotheses. Poster presented at the 20th annual CUNY Conference on Human Sentence Processing, La Jolla, CA.

The humorous "headlines" below illustrate how the same string of words can be put together into different structures with different meanings:

SISTER REUNITED AFTER FIFTEEN YEARS IN CHECKOUT COUNTER

BRITISH LEFT WAFFLES ON FALKLAND ISLANDS

STOLEN PAINTING FOUND BY TREE

What is less obvious is the fact that we are faced with thousands of "temporary" ambiguities each day in the course of processing language that we read or hear. Because the language processing system "eagerly" begins to make guesses about structure and meaning as soon as possible, sometimes incorrect guesses will be made which require later backtracking. If you were viewing the display below, you would likely experience momentary confusion about how to interpret the instruction to "Put the apple on the towel in the box." (From Spivey et al., 2002)

Work in our lab investigates what factors make resolving such ambiguities easy or difficult. By studying this, we can gain some understanding about how the language processing system is structured.

Sedivy, J. (2002). Invoking discourse-based contrast sets and resolving syntactic ambiguities. Journal of Memory and Language, 341-370.

Spivey, M., Tanenhaus, M., Eberhard, K., & Sedivy, J. (2002). Eye Movements and Spoken Language Comprehension: Effects of Visual Context on Syntactic Ambiguity Resolution. Cognitive Psychology, 45, 447-481.