Friday, 1 March 2013

Conceptual Combination: An Embodied approach


Conceptual combination (CC) is the process whereby complex concepts are constructed from simpler constituents (e.g., stone wall from 'stone' and 'wall'). Basically, it investigates the processes involved in creating and understanding new meanings from old referents. So it investigates how one may interpret novel combinations e.g. 'cactus beetle', 'elephant complaint' and so on.  The ability to construct such combinations is essential to the open-ended, creative character of human cognition, allowing the production of an unlimited set of ideas from a finite base. Researchers interested in concepts have studied at length the psychological phenomena associated with CC in order to better understand the cognitive processes underlying it. However, research on CC has thus far failed to yield any consensus on the nature of these processes. For a comprehensive and critical review of the major theories and models of conceptual combination click here.


The fact that research into this area has failed to yield  any consensus is strongly confirmed by the vast array of various theories of CC - why is this some may ask? Seeing that CC research investigates the processes involved in creating new meaning from old referents, it is pivotal that any theory of CC is able to explain this constructive ability in full and predict the resultant behavior. Up until recently, there was no CC theory that adopted an embodied approach. Did it ever dawn on previous researchers that any theory of CC that fails to take an embodied view of the conceptual system, clearly cannot account for the role of perceptual, motor, and affective information in conceptual combination? Evidently not! 

Thankfully, to address this oversight which is evident in the all other CC theories,  in 2010, Dermot Lynott & Louise Connell (both of whom completed their phd studies here at UCD) proposed the ECCo (embodied conceptual combination) model. "In ECCo, conceptual combination is the result of the interaction of the linguistic and simulation systems, such that linguistic distributional information guides or facilitates the combination process, but the new concept is fundamentally a situated, simulated entity". Take here for example the novel compound 'cactus beetle' - it is "represented as a multimodal simulation that includes visual (e.g., the shiny appearance of a beetle) and haptic (e.g., the prickliness of the cactus) information, all situated in the broader location of a desert environment under a hot sun, and with (at least for some people) an element of creepy-crawly revulsion". The ECCo theory is the only known theory that argues for a distinct role for the linguistic system during conceptual combination. 

"All conceptual combinations are situated, meaning that they involve representing a broader setting as part of the simulation. When a compound is presented, both linguistic and simulation systems are rapidly engaged; activation begins to spread out from the words to other linguistic tokens, the neural mechanisms involved in direct experience begin to simulate perceptual, motor, affective, and other situated information, and the two systems continually feed into one another (i.e., words help to activate simulations, and simulations help to activate words)."

Some evidence for Linguistic distributional information in CC: Conceptual processing uses linguistic distributional information. That is, people are sensitive to the distributional, statistical patterns in language and the wider environment, and this sensitivity provides a powerful generalized learning mechanism from early infancy. Solomon and Barsalou (2004) showed that responses in property verification tasks, where participants judge whether a property is usually true of an object (e.g.,lemon-yellow) are predominantly based on word associations (e.g., “lemon” and “yellow” are closely associated, therefore respond yes). Also, Louwerse and Jeuniaux (2010) found that word order was a significant predictor of response times when subjects were presented with two images in vertical alignment (e.g., a lamp above a table) and asked to judge whether the items usually appeared in those relative positions in the real world.

Some evidence for situated simulation information in CC: It should be noted that previous experience with language will not suffice when trying to judge whether a description of a novel situation is sensible, because for example, one's capacity to understand how a jumper filled with leaves can be used as a pillow is rooted in their ability to simulate the objects’ affordances and mesh them into a coherent situation. Also, when reading about everyday objects, people simulate perceptual properties such as shape (Zwaan et al., 2002), color (Connell, 2007; Connell and Lynott, 2009), and spatial location (Estes et al., 2008). Neuroimaging studies have also demonstrated modality-specific perceptual simulation during conceptual processing (click here for more information on Neuroimaging evidence).

Differential task demands [sensibility judgments & interpretation generation]: In a sensibility judgement task, the participant is simply asked whether a novel noun-noun compound is sensible. This is a relatively shallow judgment for which the linguistic system offers a 'quick and dirty shortcut' (heuristic). "If a compound consists of two words that have no shared statistical, distributional history, then the linguistic system will offer an heuristic for rejecting the compound as non-sensical without any attempt at conceptual combination actually taking place. On the other hand, if a compound consists of two words that are very frequently juxtaposed, then the linguistic heuristic will lead to its acceptance as sensible". As the researchers point out, obviously one does not need to rely on this linguistic shortcut and can verify compounds that seem linguistically sensible by using the simulation system - but such a shortcut is difficult to refuse! Interpretation tasks on the other hand require deeper processing in the simulation system because one has to provide an actual interpretation for a sensible noun-noun compound. Interpretation generation tasks does not allow this linguistic shortcut to suffice. 

In the ECCo model, Lynott & Connell (2010) also propose two basic interpretation types: destructive and nondestructive, which are different from the traditional property/relation based interpretation distinction and demonstrate that both the linguistic and simulation systems make complementary contributions to the time-course of successful and unsuccessful interpretation. In 2011, Lynott & Connell tested their model and the results strongly support the ECCo theory's account of conceptual combination. 

In a 2013 study by Connell & Lynott entitled ‘Flexible and fast: Linguistic shortcut affects both shallow and deep conceptual processing’ the authors employ two conceptual combination tasks ([a] sensibility judgment & [b] interpretation generation) to convey that linguistic distributional information during conceptual processing (linguistic shortcut) “extends to the processing of novel stimuli, is used in both successful and unsuccessful conceptual processing, and is evident in both shallow and deep conceptual tasks. The authors have shown that linguistic distributional frequency can predict not only the time course of successful conceptual processing (i.e., “yes” responses in sensibility judgment), but also the time course and likelihood of failure (i.e., “no” responses). So this 2013 study looks at a role for linguistic distributional information in conceptual combination BUT does not provide evidence for simulation factors during conceptual combination, as distinct from the linguistic information. So for my cognitive science project this year I am going to extend on this work by (1) replicating the linguistic distributional information finding with a new set of novel compounds & (2) employ a measure of perceptual experience (as a measure of the simulation system) to see if that makes differential predictions for the response times in the task(s).

It really is mind blowing to see just how adopting an embodied approach has really added so much to the area of conceptual combination in a way that it forces previous researchers to reevaluate their thoughts, especially seeing as this new approach (model) is reinforced with empirical evidence. Because the ability to combine concepts plays a fundamental role in diverse cognitive processes such as learning, communication, language comprehension, the composition of thoughts, and the expansion and structuring of knowledge, it will be very interesting to see how this embodied approach to CC will shed new light and add to the research on these many areas in the future. For more on the work of the above researchers click here for Lynott and click here for Connell where you will also find a list of their publications. Click here to access their embodied cognition lab. 

No comments:

Post a Comment