Intuition is slippery to define. Robin Hogarth used thoughts that “are reached with little apparent effort, and typically without conscious awareness” in Educating Intuition.  Gigerenzer and his associates call humans “homo heuristicus” and emphasize effort reduction and selective information processing.   For instance, the lexicographic heuristic has us starting by comparing alternatives on the most important dimension.  If there is a difference, we choose the best and do not seek any more information.

I find the discussion by Tilmann Betsch and Andreas Glockner in “Intuition in Judgment and Decision Making:  Extensive Thinking Without Effort” appealing.  They see heuristics as just a part of intuition.  Heuristics, largely, seem to simplify analytic thought by leaving out effort filled information processes or by reducing the amount of information considered.  Betsch and Glockner claim that “intuition is capable of dealing with complex tasks through extensive information processing without noticeable effort.”

They see intuitive processes operating without conscious control and with little potential to be accessed by introspection.  Intuitive processes  operate in parallel.  On the other hand, analysis is conducted in a serial manner that must be deliberately controlled.

Intuitive processes:

1. Are only marginally constrained by cognitive capacity.
2. Use all pieces of information that are momentarily activated from memory and salient in the environment.
3. Share with analytic processes by performing different things.
4. Handle output formation. This means combining input information and producing a judgment or preference.

These characteristics mean that intuition relies on experience.  Betsch and Glockner assert that “the stronger prior experience has been consolidated in memory, the more likely it will be activated by situational cues and, hence, feed input to intuition.  They also point to the provocative conclusion that analysis is for input and intuition is for output.

Betsch and Glockner performed a series of experiments that showed that information integration can occur even if cognitive capacity is constrained by other tasks and even if the participants are asked to perform unexpected tasks.  As a distracting task that was framed as the primary one, participants were asked to memorize the content of video ads while fictitious stock quotes scrolled by on the bottom.  The screenshot above provides an example. While watching the videos for up to 10 minutes, the participants were asked to read the stock information aloud. After being asked to recall information about the video ads, they were unexpectedly asked to adjust a scroll bar for each of the 8 stocks indicating how much they liked it with endpoints labeled good and bad. These evaluations reproduced the actual variation in stock gains. In other words. the participants liked the stocks most that gained the most and in the same proportion as the gains. They concluded after ruling out alternative explanations that mere encoding (by reading aloud) the information started integration of the information without any intention.

In another series of studies  Betsch & Glockner considered information and integration under time constraints.  They found that information integration can be performed intuitively without conscious, weighting and adding rules. They substantiated in further experiments a connectionist model of information integration. In this model intuition processes are sensitive to the holistic aspect of the information sample, the coherence in the pattern. In coherent patterns, all the arguments point to one choice. If there are pros and cons, the choice is more difficult.  The model asserts that the more coherent the evidence, the faster a judgment can be made. Moreover, the model predicts that if additional information makes the pattern more coherent, decision time will be reduced even though the amount of information to be considered is greater.

These arguments run counter to the dominant view in JDM research, believing that capacity constraints require simplified strategies for decision makers.  However, Betsch and Glockner make three assumptions to make their findings consistent with the bounded rationality approach.

1. Intuitive and analytic processes are distinct.  Intuition is not a shortcut to deliberation.  Heuristics are, in the main, not intuitive strategies, but strategies to simplify analysis.
2. Intuitive and analytic processes have different potentials and constraints. Analysis is performed in step by step fashion and this take processing time and is governed by working memory constraints and environmental factors. Intuition seems to be little constrained by information and task constraints, but is restrained by a reduced degree of coherence of the encoded information.  Intuition can also be impaired by trying to apply analysis to it.
3. Intuition and analysis are component processes rather than different modes of thinking.

Betsch and Glockner believe that information integration and output formation (choice, preference) is intuitive.  Analysis involves directed search (looking for valid cues or asking an expert for advice), making sense of information, anticipating future events, etc.  Thus, they see a judgment as a collaboration of intuition and analysis. The depth of analysis varies, but intuition is always working so preferences are formed even without intention.  Limiting processing time and capacity constrains only input.  Thus, once information is in the system, intuition will use that information irrespective of amount and capacity.

Betsch, T & Glockner A (2010) Intuition in Judgment and Decision Making: Extensive Thinking without Effort. Psychological Inquiry, 21: 279-294.