Processing of Garden Path Jokes: Theoretical Concepts and Empirical Correlates
Bastian Mayerhofer
Cumulative thesis
Date of Examination:
2014-03-04
Date of issue:
2015-02-20
Advisor:
Schacht, Annekathrin Prof. Dr.
Referee:
Schacht, Annekathrin, Prof. Dr.
Referee:
Rakoczy, Hannes Prof. Dr.
Persistent Address: http://hdl.handle.net/11858/00-1735-0000-0022-5DC9-4
Files in this item
Name:
thesisSUBNewUpload.pdf
Size:
6,2 MB
Format:
PDF
The following license files are associated with this item:
Abstract
English
In garden path jokes (GP jokes), an initially dominant semantic representation of an ambiguous text is violated. It needs to be revised in order to re-establish a coherent representation. The processing of GP jokes is modeled on a theoretical level as a dynamic interaction between a stimulus and a recipient. Both sides of the interaction are influenced by a wide range of factors that together determine the comprehension process. The focus is put on three aspects: (i) the salience of the first interpretation, (ii) the accessibility of the hidden interpretation, and (iii) the humorous potential of the whole joke. The comprehension process is assumed as probabilistic, non-monotonic, and incremental reasoning towards the most plausible interpretation of both linguistic and non-linguistic input. Basic assumptions about the sequential nature of the processing – detection of incoherence, revision of the semantic representation, and emotional reaction (mirth) – are empirically investigated in a first series of experiments. 48 GP jokes were manipulated and presented to participants in three conditions: (i) with coherent ending, (ii) with joke ending, or (iii) with discourse-incoherent ending. A rating study (N=69), a reading times study (N =24), and three studies with recordings of ERP and pupil changes (N = 21, 24, and 24) supported the hypothesized cognitive processes. Jokes were rated as more funny, moderately unpredictable, and comprehensible. They showed increased reading times of the final word compared to coherent endings. ERP data mainly revealed semantic integration difficulties (N400). Some weak and mixed evidence is presented for an enhanced late frontal positivity (fP600), possibly related to the emotional outcome, and for a late left anterior negativity (LLAN), possibly related to increased processing effort. The emotional reaction to GP jokes is supported by larger pupil dilations in response to joke endings. Incoherent endings were rated as less funny, less predictable, and incomprehensible. They also showed increased reading times. The N400 was stronger and sustained. Incoherent endings also showed a context-sensitive P600 effect. An additional self-paced reading time experiment highlighted the influence of contextual constraint on the interpretation and re-interpretation processes by semantic priming prior to the GP joke. These findings support the theoretical conceptualization of salience and accessibility. Taken together, the processing of GP jokes is presented as playful communication that allows the strengthening of important cognitive skills related to the adaptation to changing environments. Mirth as the emotional reaction accompanying the cognitive processing is discussed as a reward for the engagement in this behavior.
Keywords: Humor; Text comprehension; Pupil dilation; ERP; N400; Garden Path Joke
