Ofer Fein

On understanding familiar and less-familiar figurative language☆

Abstract Findings of three experiments are consistent with the graded salience hypothesis (Giora, 1997), according to which salient meanings should be processed initially before less salient meanings are activated. A meaning of a word or an expression is considered salient if it can be retrieved directly from the mental lexicon. According to the graded salience hypothesis, processing familiar metaphors (which have at least two salient interpretations — the literal and the metaphoric) should involve activation of both their metaphoric and literal meanings, regardless of the type of context in which they are embedded. Processing less familiar metaphors (which have only one salient meaning — the literal) should activate the literal meaning in both types of contexts; however, in the literally biased context, it should be the only one activated. Processing familiar idioms in a context biased towards the idiomatic meaning should evoke their figurative meaning almost exclusively, because their figurative meaning is much more salient than their literal meaning. However, processing less familiar idioms in an idiomatic context should activate both their literal and idiomatic meanings, because both meanings enjoy similar salience status. In a literally biased context, familiar idioms should evoke their more salient idiomatic meaning to a greater extent than less familiar idioms. A word fragment completion test was used to measure the amount of activation of literal and figurative meanings in both literally and figuratively biased contexts. Subjects were presented with ‘target sentences’ (metaphors or idioms) at the end of either figuratively or literally biased contexts. They were asked to complete fragmented words (such as t_b_e) with the first word that came to mind. The target words were related to either the figurative or the literal meaning of the target sentence, so that activation of the different meanings could be assessed. Findings reveal that, contrary to current beliefs, metaphor interpretation involves processing the literal meaning. They further reveal that metaphor and literal interpretations do not involve equivalent processes.

Irony: Context and Salience

Two experiments test a graded salience account of irony processing (Giora, Fein, & Schwartz, 1998). Experiment 1 shows that, as predicted, less familiar targets embedded in ironically biasing contexts facilitate only the salient literal meaning initially: 150 msec after their offset. However, 1,000 msec after their offset, the less salient ironic meaning becomes available and the literal meaning is still as active. In contrast, familiar ironies facilitate both their salient literal and ironic meanings initially: 150 msec after their offset. Results do not change significantly after a 1,000-msec delay. In the literally biasing contexts, less familiar ironies facilitate only the salient literal meaning. In contrast, familiar ironies facilitate both their salient literal and ironic meanings under both interstimulus interval conditions, as predicted. Experiment 2 confirms that these findings were affected by the target sentences rather than by the contexts themselves. In Experiment 2, the contexts were presen...

Weapons of Mass Distraction: Optimal Innovation and Pleasure Ratings

In 6 experiments we test the Optimal Innovation Hypothesis, according to which an optimally innovative stimulus, such that induces a novel response while allowing for the recovery of a salient one (Giora, 1997b, 2003), would be rated as more pleasing than either a more or a less familiar stimulus. Experiment 1 shows that it is the stimulus that meets the requirements for optimal innovativeness that is most pleasurable. Reading times obtained in Experiment 2 support the assumption that the stimuli found most pleasurable involve processing a salient meaning. Experiment 3 corroborates the results of Experiment 1 and Experiment 2, showing that they also hold for identical (rather than different) stimuli. Experiment 4 controls for the possibility that the lengthy reading times found earlier might reflect lack of understanding. Experiment 5 shows that optimal innovation supersedes figurativity. Experiment 6 demonstrates that the Optimal Innovation Hypothesis applies to nonverbal stimuli as well.

Salience and Context Effects: Two Are Better Than One

This study provides evidence supporting the hypothesis that language comprehension involves 2 separate mechanisms that run in parallel: a linguistic mechanism and a contextual mechanism. The linguistic mechanism (e.g., lexical access) is modular and stimulus driven; it is a bottom-up, perceptual mechanism, induced by a lexical stimulus to search the mental lexicon for its match. This mechanism is encapsulated with respect to nonlinguistic information and thus operates locally (i.e., on the word level). Lexical access is exhaustive and ordered: Salient meanings are accessed faster. Contextual facilitation, on the other hand, is the outcome of a central, expectation-driven mechanism that operates globally during language comprehension at the point where prior linguistic information has already been processed and interfaced with other cognitive processes (e.g., inferencing). Experiment 1 indicates that contextual facilitation can occur even before lexical accessing takes place, fostering an impression of a selective process. Experiment 2 shows that the target words position in the sentence (initial vs. noninitial) is crucial for the operation of the predictive mechanism. Thus, we would not expect contextual meanings to outweigh salient meanings at the beginning of sentences.

Expecting Irony: Context Versus Salience-Based Effects

Results from 4 experiments support the view that, regardless of contextual information, when an end-product interpretation of an utterance does not rely on the salient (lexicalized and prominent) meanings of its components, it will not be faster than nor as fast to derive as when it does. To test this view, we looked into interpretations of salience-based (here, literal) interpretations and expectation-based (here, ironic) interpretations in contexts inducing an expectation for irony. In Experiment 1, expectancy was manipulated by introducing an ironic speaker in vivo who also uttered the target utterance. Findings show that ironic targets were slower to read than literal counterparts. Experiment 2 shows that ironies took longer to read than literals and that response times to ironically related probes were longer than to literally related probes, regardless of context. Experiments 3 and 4 show that, even when participants were given extra processing time and were exclusively presented ironically biasing contexts, the expectancy for irony acquired throughout such exposure did not facilitate expectancy-based compared to salience-based interpretations.

Salience and Context: Interpretation of Metaphorical and Literal Language by Young Adults Diagnosed with Asperger's Syndrome

Aspergers Syndrome (AS) involves difficulties in social communication but no delays in language or cognitive development. According to the received view, individuals with AS are biased toward the literal and are insensitive to contextual cues. According to the graded salience hypothesis (Giora, 1997, 2003), participants with AS and controls would be sensitive to both context and degree of salience rather than to degree of nonliterality. Our results show that while individuals with AS generally performed worse than controls, their overall pattern of response was similar to that of controls: both groups performed worse on novel than on familiar expressions, whether literal or metaphorical; both groups benefited from context, which reduced response times and error rates on novel but not on familiar metaphors; both groups rated negative utterances as more metaphoric than their affirmative counterparts. Individuals with AS, then, are sensitive to context and degree of salience and are not biased toward the literal.

Defaultness Reigns: The Case of Sarcasm

Findings from two experiments (run in Hebrew) argue in favor of the superiority of default, preferred interpretations over non-default less favored counterparts, outshining degree of (a) non-salience, (b) non-literalness, (c) contextual strength, and (d) negation. They show that, outside of a specific context, the default interpretation of specific negative constructions (He is not the most organized student) is a non-salient interpretation (here sarcastic)1; their non-default interpretation is a salience-based alternative (here literal). In contrast, the default interpretation of the affirmative counterparts (He is the most organized student) is a salience-based interpretation (here literal); their non-default interpretation is a non-salient alternative (here sarcastic; Experiment 1). When in equally strongly supportive contexts, default yet non-salient negative sarcasm is processed faster than (1) non-default, non-salient yet affirmative sarcasm and (2) faster than non-default yet salience-based negative literalness. Complementarily, default, salience-based affirmative literalness is derived faster than (1) non-default non-salient affirmative sarcasm, and (2) faster than non-default albeit salience-based negative literalness (Experiment 2). This unparalleled quadrilateral pattern of comparisons speaks to the superiority of defaultness.

Contextual Strength: the Whens and Hows of Context Effects

Highlighting the role context plays in shaping our linguistic behaviour is the major contribution of pragmatics to language research. Indeed, pragmatics has shifted the focus of research from the code to contextual inference (Carston, 2002; Sperber & Wilson, 1986/1995). It is widely agreed now that contextual information is a crucial factor determining how we make sense of utterances. The role of context is even more pronounced within a framework that assumes that the code is underspecified allowing for top-down inferential processes to narrow meanings down and adjust them to the specific context.

On the priority of salience-based interpretations: The case of sarcastic irony

Abstract Results from 2 experiments support the view that, regardless of strength of contextual expectation for utterance nonsalient (ironic) interpretation, (a) salience-based interpretations will not be blocked. Instead, they will be facilitated initially. And, (b) if conducive to the interpretation process, they will not be suppressed, albeit incompatible (Giora 2003; Giora and Fein 1999a; Giora and Fein 1999b; Giora et al. 2007). In Experiment 1, expectancy for an ironic utterance was manipulated by introducing an ironic speaker, whose ironic utterances were prefaced by overt ironic cues, making explicit the speakers ironic intent. In Experiment 2, expectancy strongly biased via repeated exposure to ironic utterances, was further strengthened by informing participants that the experiment was testing sarcasm interpretation. Long processing times were allowed so as to tap later (suppression) processes. Results from reading times and lexical decisions support the temporal priority of salience-based interpretations, while arguing against both, the contextualist views (Gibbs 2002; Katz 2009) and the Gricean suppression hypothesis (Grice 1975).

Negation Generates Nonliteral Interpretations by Default

Four experiments and 2 corpus-based studies demonstrate that negation is a determinant factor affecting novel nonliteral utterance-interpretation by default. For a nonliteral utterance-interpretation to be favored by default, utterances should be potentially ambiguous between literal and nonliteral interpretations. They should therefore be (a) unfamiliar, (b) free of semantic anomaly or any kind of internal incongruity, and (c) unbiased by contextual information. Experiments 1–3 demonstrate that negative utterances, meeting these 3 conditions, were interpreted metaphorically (This is not a safe) or sarcastically (Ambitious she is not) when presented in isolation and were therefore processed faster in contexts strongly biasing them toward their nonliteral than toward their (equally biased) literal interpretation. Experiment 4 reduces the possibility that it is structural markedness on its own that induces nonliteralness. Two corpus-based studies provide corroborating evidence, supporting the view of negation as an operator generating nonliteral interpretations by default.