Albert Katz – A Potpourri of Recent Research from the Laboratory
Department of Psychology
University of Western Ontario
For over 35 years, I have been engaged in understanding how one translates the surface appearances of objects and events into deeper psychological understanding. Trained as a cognitive psychologist, this interest has led me to study autobiographical memory and the processing of language when one says something that, on the surface, differs from what one is attempting to convey, such as when one uses metaphor, sarcasm, proverbs or puns. As an experimental psychologist, my students and I have attempted to bring real-life phenomena into the laboratory. Here I discuss two lines of related research that have emerged from my laboratory, with special emphasis on our most recent findings.
The mechanisms behind deeper understanding
This line of research is aimed at identifying the mechanism (s) underlying the translation of surface expressions into deeper understanding. One component examines the semantic organization of knowledge that gets activated on hearing or reading a sentence. For instance, consider a simple nominal metaphor such as “my surgeon is a butcher”. The concept being described (the target “surgeon”) and the concept providing information about the target (the source “butcher”) can be considered as residing in an n-dimensional semantic or similarity space, such that each of these concepts can be “close” to many or few related concepts. We are using this basic framework to understand questions such as why some metaphors are easier to understand than others or why sometimes one usually cannot reverse the concepts (e.g., “my butcher is a surgeon”) by either changing the nature of the interpretation, or more commonly, producing a nonsensical statement.
James Boylan, a PhD student in my laboratory is interested in understanding the nature of humor evoked in language. He employs puns as his target stimuli and has developed a novel means of determining the semantic similarity of alternate senses implied in puns using dictionary meanings. He has demonstrated that the greater the semantic distance (or incongruity) between the alternate meanings the more the pun is perceived as funny. Moreover, he has shown that with increasing repetition of the pun it becomes less funny. Importantly, the semantic distance between the implicit alternate meanings activated by a pun determines the extent to which the humor associated with a pun decreases with repetition (see Boylan & Katz, 2015).
Another PhD student in the laboratory, Hamad Al-Azary is also examining semantic effects but in his case he studies metaphor, taking a decidedly experimental approach. He has completed a series of studies that started with his MA work in Lori Buchanan’s lab at the University of Windsor. In his approach, he creates nominal metaphors (e.g., “love is a ladder”) by systematically varying semantic factors (such as semantic density: the number of related concepts in semantic space to the topic and source concepts) and concept concreteness. In recent studies in our lab (Al-Azary & Katz, 2015a, b) we have examined the comprehension of novel metaphoric expressions in which word concreteness, semantic density and embodiment of the vehicle have been manipulated. I will discuss embodiment presently. Here embodiment was operationalized by the degree of rated Body-Object Interaction (BOI). Body-object interaction assesses the ease with which a human body can physically interact with a word’s referent. These studies are ongoing, with the data to date suggesting that density in semantic space and concreteness of the words expressed in the metaphor both play a role in comprehending a metaphor and suggestively, in the degree to which the source and vehicle can be interchanged (e.g., “education is a ladder” versus “a ladder is education”). The effects of BOI are negligible suggesting that any role played by this one measure of embodiment is at best secondary to semantic factors in determining how we understand novel nominal metaphors.
The second processing component we have been studying within this line of research arises from the notion that bodily reactions serve a central and reciprocal role in the interaction of language and meaning. As extended to the study of non-literal language this position suggests that a bitter reaction to let’s say a certain type of food would influence us to interpret a written or spoken language as sarcasm (rather than as sincere) and reading a sarcastic passage would bias a person to perceive a bland food as more bitter. My colleague Dr. Hussey and I are currently engaged in such studies. Some of my students are engaged in other aspects of the embodiment position.
My then-PhD student, Dr. Andrea Bowes examined the finding that the processing of metaphor creates a sense of interpersonal closeness. She reasoned this sense of embodied emotional closeness might be reflected in other ostensibly untreated tasks, dealing with the perception of emotion in others. In her studies the unrelated task was the Reading the Mind in the Eyes Test (RMET), a test of theory of mind often used in studying autistic populations. In the RMET one is asked to identify the emotions expressed in pictures of eyes and the just surrounding portion of the face. As predicted, those people who read a set of metaphors prior to completing the RMET were more accurate in identifying emotions than those who read as set of non-metaphor counterparts; even in one case when the sentences were presented without any discourse context (Study 3; see Bowes & Katz, 2015).
Nick Reid, just completing his MA, has studied an embodiment explanation further by replicating the reading portion of the third study in Bowes and Katz study but having the unrelated task to be the judgment of the distance between two people in a set of pictures. Over several studies he has found that those who read metaphor judged the people as standing further apart than those who read the non-literal sentence controls. In follow up studies he discounts perceived intimacy as a factor leading to this effect and is now examining whether the semantic distance between topic and source in the metaphor projects into the perception of physical distance between people.
Finally, my MA student, Krysten Zator has found evidence that cueing autobiographical memory by music (versus verbal cues) evokes substantially more memories described with spatial and movement imagery and is now directly testing embodiment by examining the effects of motor activity on musicians (versus non-musicians) when listening to short musical phrases.
The social impact of non-literal language
The second line of research is directed towards analyzing the social impact of using metaphor and other forms of non-literal language. The Bowes and Katz (2015) study and the work of Nick Reid described above indicates two ways in which the processing of metaphor has unexpected social impact: sensitizing people to the emotional state of other people and in effecting the perception of the physical distance between two people. In our earliest attempts to examine social effects we used standard experimental procedures employed in cognitive psychology, such as embedding metaphors within a passage and asking questions about the communication goals being met, or by examining online the speed with which different sentences are read. The novel aspect was the manipulation of socially relevant information in the passage, such as reference to occupation or social class. We found that the mere presence of this information biased the way people interpreted a metaphor (e.g., “children are such gems”), with some the mention of occupations associated with “white color” occupations (e.g., professor) inviting a pure metaphoric interpretation whereas the mention of “blue color” occupations (e.g., truck driver) biasing a person to perceive the metaphor as being used as sarcastic irony (see Katz & Pexman, 1997). We showed further this effect occurred online, biasing interpretation during the act of reading the metaphor (Pexman, Ferretti & Katz, 2000).
We have shown, as well, gender and friendship effects. Using chat rooms we had same-gender dyads communicate persuasively on topics of common interest. In one study, one of the interlocutors had to try to persuade the other (for instance in taking a specific course at university) and in another study the interlocutors worked collaboratively in producing arguments that could be used to convince a third party. We subsequently analyzed what the people communicated with one another finding that males used reliably more metaphor in their persuasive arguments overall, whether communicating with a friend or a stranger. Females used much fewer metaphors overall and virtually none when talking to a stranger (see Hussey & Katz, 2006). We examined further whether metaphor usage is a cue to identity, finding that the mere use of metaphor in written text biases the reader to see the metaphor-user as male or masculine (Hussey & Katz, 2009).
Finally, we have started to directly look at the discourse contexts in which people use metaphor or irony. In Bowes and Katz (2015, study 2) participants were presented either a set of metaphors (e.g., “the woman dove into her knitting”) or controlled literal statements differing by one word (e.g., “the woman dove into her swimming pool”). Each group was asked simply to provide a passage in which the presented sentence would be used and the generated contexts were then analyzed. The contexts generated to the metaphor differed by reliably greater use of emotional idioms and cognitive mechanism terms (words such as “think”, “feel”, “intend”) from those generated to the literal statements metaphor, indicating the greater care in how one constructs the ecology in which to present metaphor to make it understandable to other people.
My former PhD student Dr. John Campbell and I employed a similar task to test various theories of sarcastic irony. In Campbell and Katz (2012) we manipulated instructions such that half of our sample of students was asked to produce a discourse such that the ambiguous statement (e.g., “You a good friend”) would be understood as sarcasm whereas the other half of our sample, given the exact same ambiguous statements, were asked to generate a discourse context in which the statement would be understood as sincere. The contexts so generated were then subjected to analyses to identify how the discourse differed, and were shown to differ from one another along the dimensions presumed as necessary to sarcastic irony by various theories (failed expectation, pragmatic insincerity, negative tension, and presence of a victim) and along stylistic components (as indexed by the Linguistic Inquiry and Word Count program). However, none of these components were found to be necessary. Indeed, in each case, the items rated as highest in sarcasm were often at the lowest levels on the putative “necessary” characteristic. These data are taken as consistent with constraint satisfaction models of sarcasm processing in which various linguistic and extra-linguistic information provide probabilistic (but not necessary) support for or against a sarcastic interpretation. We believe that the social cues identified and discussed above (e.g., gender, social class) and the means in which use of non-literal language is framed (e.g., use of cognitive content words, presence of a specific victim or target of a barb, choice of a source item from a specific portion of semantic space) are among the sources of probabilistic information used to convey intent and permit comprehension of that intent.
Al-Azary, H., & Katz, A. N. (2015a). Novel Metaphor Comprehension: A Disembodied Advantage. Paper presented at the Fifth International Conference on Metaphor in Language and Thought in Belo Horizonte, MG, Brazil. October 7-9.
Al-Azary, H., & Katz, A. N. (2015b). Novel Metaphor Comprehension: Topic Concreteness, Semantic Neighborhood Density, and Directionality. Poster presented at the Annual Meeting of the Psychonomic Society in Chicago, IL, USA. November 19-22.
Bowes, A., & Katz, A. (2015). Metaphor creates intimacy and enhances one’s ability to infer the internal states of others. Memory and Cognition, 43, 953-963.
Boylan, J., & Katz, A. (2015). Semantic Incongruity and Humour in Written Puns. Poster presented at the Annual Meeting of the Psychonomics Society in Chicago, IL, USA. November 19-22.
Campbell, J., & Katz, A. N. (2012). Are there necessary and sufficient conditions for inducing a sense of sarcasm? Discourse Processes, 49, 459-480.
Hussey, K., & Katz, A. N. (2006). Metaphor production in online conversation: Gender and friendship status. Discourse Processes, 42, 75-98.
Hussey, K., & Katz, A. N. (2009). Perception of the use of metaphor by an interlocutor in discourse. Metaphor and Symbol, 24, 203-236.
Katz, A. N., & Pexman, P. (1997). Processing of figurative language: Occupation of speaker turns metaphor into irony. Metaphor and Symbolic Activity, 12, 19-41.
Pexman, P., Ferretti, T., & Katz, A. (2000). Discourse factors that influence on-line reading of metaphor and irony. Discourse Processes, 29, 201-222.