Symbolically speaking: a connectionist model of sentence production

The ability to combine words into novel sentences has been used to argue that humans have symbolic language production abilities. Critiques of connectionist models of language often center on the inability of these models to generalize symbolically (Fodor & Pylyshyn, 1988; Marcus, 1998). To address these issues, a connectionist model of sentence production was developed. The model had variables (role‐concept bindings) that were inspired by spatial representations (Landau & Jackendoff, 1993). In order to take advantage of these variables, a novel dual‐pathway architecture with event semantics is proposed and shown to be better at symbolic generalization than several variants. This architecture has one pathway for mapping message content to words and a separate pathway that enforces sequencing constraints. Analysis of the model's hidden units demonstrated that the model learned different types of information in each pathway, and that the model's compositional behavior arose from the combination of these two pathways. The model's ability to balance symbolic and statistical behavior in syntax acquisition and to model aphasic double dissociations provided independent support for the dual‐pathway architecture.     

Face recognition algorithms and the other‐race effect: computational mechanisms for a developmental contact hypothesis

People recognize faces of their own race more accurately than faces of other races. The “contact” hypothesis suggests that this “other‐race effect” occurs as a result of the greater experience we have with own‐ versus other‐race faces. The computational mechanisms that may underlie different versions of the contact hypothesis were explored in this study. We replicated the other‐race effect with human participants and evaluated four classes of computational face recognition algorithms for the presence of an other‐race effect. Consistent with the predictions of a developmental contact hypothesis, “experience‐based models” demonstrated an other‐race effect only when the representational system was developed through experience that warped the perceptual space in a way that was sensitive to the overall structure of the model's experience with faces of different races. When the model's representation relied on a feature set optimized to encode the information in the learned faces, experience‐based algorithms recognized minority‐race faces more accurately than majority‐race faces. The results suggest a developmental learning process that warps the perceptual space to enhance the encoding of distinctions relevant for own‐race faces. This feature space limits the quality of face representations for other‐race faces.     

Why John married Mary: Understanding stories involving recurring goals

A story understander needs a great deal of knowledge about people's goals. This knowledge is needed to infer explanations for the behavior of the characters in a story. Sometimes a character's behavior cannot be explained in terms of a particular goal, but only in relation to a set of recurring goals that a character anticipates having. The concept of goal subsumption is introduced to deal with these situations. Goal subsumption is a way of planning for many goals at the same time. It occurs in three different types of story situations, each of which has its own rules for recognition and understanding.     

Decision theory and artificial intelligence II: The hungry monkey

This paper describes a problem-solving framework In which aspects of mathematical decision theory are incorporated into symbolic problem-solving techniques currently predominant in artificial intelligence. The utility function of decision theory IS used to reveal tradeoffs among competing strategies for achieving various goals, taking into account such factors as reliability, the complexity of steps in the strategy, and the value of the goal. The utility function on strategies can therefore be used as a guide when searching for good strategies. It is also used to formulate solutions to the problems of how to acquire a world model, how much planning effort is worthwhile, and whether verification tests should be performed. These techniques are illustrated by application to the classic monkey and bananas problem.     

Passing markers: A theory of contextual influence in language comprehension

Most Artificial Intelligence theories of language either assume a syntactic component which serves as “front end” for the rest of the system, or else reject all attempts at distinguishing modules within the comprehension system. In this paper we will present an alternative which, while keeping modularity, will account for several puzzles for typical “syntax first” theories. The major addition to this theory is a “marker passing” (or “spreading activation”) component, which operates in parallel to the normal syntactic component.     

The role of affect in narratives

Affect states and reactions commonly occur in narratives. This paper discusses the importance of affect knowledge and processing in the context of BORIS, a computer program which reads and answers questions about narratives involving multiple sources of knowledge. There are several reasons why affects are important in a process model of narrative comprehension: For one thing, affects describe goal situations and signal the occurrence of expectation failures. Affective reactions also serve as an indication of the status of interpersonal relationships. Finally, affects influence the kinds of thematic structures which become instantiated in episodic memory.