P335 Cognitive Psychology, Prof. Kruschke

Exam 1 Questions from Students (with answers)

Go to our class discussion group to read and post Q&A about Exam 1.

Below is a set of Q&A from previous classes, many of which are still relevant. But please browse our current discussion page first! To find a particular topic, scan the list below, and also search this document for specific keywords that interest you by using the FIND button on your web browser. (The FIND function is also located under the Edit menu.)
Questions regarding:
Representation and process.
Evaluating feature theories.
Face priming and interactive-activation model.
Visual search and parallel/serial processing.
Too much! How can I learn it all?
What's in a theory? (representation and process)
Interactive activation model.
Illusory conjunctions.
Attention to events (Neisser & Becklen experiment).
Pre-attentive and attentive levels of processing.
Filter theory (Broadbent's attention theory).

(Two different people sent these questions:)
> Professor Kruschke could you please explain representation and
> process to me.  I don't really understand how it is used and how we
> determine what it is in an experiment.  Thanks

> How are you?  I have hard time distinguish between the meaning of
> process and representation.  I cannot tell the difference between
> them in an experiment.

First, representation and process are in a theory or explanation of experimental results, not in an experimental design or effect.

In an explanation of how or why an effect occurs, the explanation makes claims about what information is used in the mind, and how that information is used. The representation in a theory is what information is used. The process in a theory is how the information is used.

For example, in a feature theory of pattern recognition, the represention of patterns is by features (formatted as an unordered set), and the process of recognizing patterns is checking for which features are shared by the stimulus and the patterns in memory.

See also other related questions below.

> I'm unclear about the procedures for evaluating feature theories of
> pattern recognition.  As far as Gibson's experiment and theory, can
> you give me an idea of what info is needed in order to be thorough
> but not too broad?  Thanks a lot

The basic idea is that a detailed feature theory has to specify just which features are actually used in human pattern perception. Once a theorist hypothesizes a set of features as being the correct ones, then the feature theory needs to be tested, i.e., evaluated. How can we tell if the hypothesized features are really used in human perception? The textbook (starting on p.25, with a section header, Evaluating feature theories) discusses a few methods, such as identification confusion errors, response times for distinguishing patterns, and threshold identification accuracy.

To adequately describe Gibson's approach, you need to fill out the four leaves of the knowledge organization diagram: (1) Stimuli, task, procedure, I.V.'s and D.V.'s; (2) Results; (3) Representation; and, (4) Process. The book only gives the key elements of each of these parts, so it would be difficult to be too thorough with this one. See other questions and answers for a little more discussion of representation and process.

(Two different people sent these questions:)
> Hello again.  How does the interactive-activation model explain face
> priming?  Thanks for your time!

> Help me to understand what you would like me to focus on when I am
> studying the word-superiority effect and the interactive-activation
> model....and also how it deals with face-priming.

A photo of Nixon primes recognition of a photo of Kissinger. The interactive-activation model explains this by hypothesizing top-down feedback from semantic associations of Nixon with Kissinger. Thus, seeing the photo of Nixon activates knowledge of Kissinger which feeds back down to prime the nodes which recognize Kissinger.

This explanation is the same kind as that for the word-superiority effect: The key aspect of the explanation is the top-down feedback from (previously learned) knowledge.

There is no priming of Elvis from Nixon because, for most people, there is little association from Nixon to Elvis. A similar lack of priming occurs in the interactive-activation model of the word-superiority effect: When a non-word is presented, there is no association in word knowledge to feedback to letter recognition.

See other questions and answers about the interactive-activation model.

> Dr. Kruschke: I am having a trouble finding information on parallel
> and serial search processes in visual search tasks.  Can you tell me
> what I need to know and where to find it.  Thanks

The first homework assignment has info about parallel and serial processing, and the lecture the day is was due gave information about visual search experiments. Compare notes with a classmate. Visual search experiments and parallel or serial processing theories are not discussed in our textbook. Many other cognitive psychology textbooks do describe them, however (take a look in the library). The article cited at the end of the homework also is short and readable and full of info.

So what should you know? You should be able to fill in the four leaves of the knowledge organization diagram: (1) Stimuli, task, procedure, I.V.'s and D.V.'s; (2) Results; (3) Representation; and, (4) Process. For visual search, two I.V.'s were number of items and presence/absence of the target. A third I.V. was whether the target was defined by a unique feature relative to the distractors, or only by a conjunction of features. The results for feature search could be accounted for by parallel processing, and the results for conjunction search could be accounted for by serial processing. As stated in the homework, these theories emphasized different processes, and didn't say much about the representation. The representation consists of pre-attentively registered features, and post-attentively recognized patterns.

>       I'm a student in your P335 class and I'm beginning to
> study for the upcoming exam.  The thing is I'm having great
> difficulty trying to piece together information to fully
> answer your example topics.  I don't really know how to
> tackle them.  When I look at the readings my mind goes
> bonkers because I feel that the readings are fairly difficult
> to understand and very technical.  Should I be focusing on
> the notes more than the book or both.  I need help!!!!
> Thanks.
>  Sincerely, Confused

Keeping your knowledge organized is crucial for understanding and for remembering. The time you spend figuring out how it all fits together will pay off in terms of performance on exams and in terms of your enjoyment of the class!

To help digest the readings, it's crucial always to keep in mind the "big picture" --- why is the author discussing this particular thing in this particular place? What point is it supposed to be illustrating or demonstrating? Be sure you can answer this question; if you can't, then you don't really understand!

The knowledge organization diagram from the syllabus is intended to help you keep the big picture. Whenever you read about an experiment in the book, or hear about one in lecture, you should be able to fill in all the leaves of the knowledge organization tree. Sometimes an experment is discussed in terms of its results only, without a theory thoroughly described. In these cases, you should at least be aware that if a theory were proposed, it would have to specify representation and process.

> "what is in any theory of a cognitive phenomenom?" has me a
> little confused.  I don't exactly know what you mean by this
> question.  Thank you for your time!
> I am in the process of studying for the exam we have on
> Thursday, and I am afraid I am having trouble deciding what
> is in a theory of cognitive phenomenon.  Any insight you
> could give me as to what specifically you are asking would
> be very helpful.  Thank you.

Any theory of a cognitive phenomenon must specify the representation of information and the processes that act on it. The representation includes the content of the information, i.e., what is represented, and the format or organization of the information.

Example: In finding out a phone number, the input is a person's name and the output is his or her phone number. One theory of how this is accomplished claims that the information represented in memory is a set of names with corresponding phone numbers (that's the content), arranged alphabetically by last name (that's the format). When a name is presented, the memory is searched by starting in the list approximately where names with the corresponding first letter are stored, and then scanning forward or back toward the correct position in the list (that's the process). In class I also presented an alternative theory of finding out a phone number.

Several other examples were presented in the topic of pattern recognition; e.g., the interactive activation model of the word-superiority effect.

> Could you please explain what the representation and process
> are in the interactive activation model? I think I can figure
> it out, but I want to make sure. Thanks!

In general, representation consists of what information is there (contents) and how it is formatted or organized. The interactive activation model represents the following information (contents): features (of letters), letters, and words. This information is organized or formatted into three corresponding levels.

The processing in the model is spreading activation, from features to letters to words. Crucially, to account for the word-superiority effect, there is also spreading activation from the words back down to the letters.

>      Could you please explain illusory conjunctions, and what
> we should know about them for the exam?  I couldn't find them
> in the book, and my notes were insufficeint.  Thank you for
> your time.

An illusory conjunction is the perception of a combination of features that are not really combined in the world. For example, seeing a red A and a green B where there is really a green A and a red B.

Illusory conjunctions are predicted and explained by Treisman's Feature Integration Theory, in which features are pre-attentively registered, but then correctly combined only with subsequent attention. If attention is distracted, illusory conjunctions might (and sometimes do) occur.

Notice that these ideas -- about the role of attention -- fit nicely with Sperling's (and other's) ideas about iconic memory, and Rock & Gutman's ideas about attention and shape perception.

> Are you going to discuss in more detail the Neiser and
> Becklen experiment since we are responsible for the
> information on the exam and because it is not discussed in
> the book? I did not get enough information about it.  Thank
> you.

The point of the Neisser and Becklen experiment was simply that we can selectively attend to whole events (such as games), even when we are looking at both events simultaneously. In fact, it is very difficult to divide attention across both events, even though both events are visually in the same place.

This happens in everyday life when we look out a window during twilight, and we can see outside or the reflection of inside merely by shifting attention from one event to the other. The reason for doing the experiment was so this everyday phenomenon could be carefully controlled and measured, to be sure that what we merely feel in everyday life really is how we perform when various extraneous influences are eliminated.

The overall point of my discussion of attention to locations, objects and events was that we do not attend only to locations, but also to objects and to whole events. We do not attend only to events or only to objects; we can attend at all three levels. Indeed, attention occurs at every level of processing in perception and cognition.

> I'm having trouble trying to figure out what exactly you
> wnat us to know about pre-attentive and attentive levels
> of perception?  Thanks.

First, know the difference. Various theories assert that some information is extracted or registered pre-attentively, but other types of information need attentive processing.

Second, know which theories rely on the distinction. This distinction shows up in theories of results from the partial report procedure (Sperling, Rumelhart), perception of shape (Rock & Gutman), illusory conjunctions (Treisman), dichotic listening (Broadbent, Treisman, et al.), etc.

Third, know that it is really just a theoretical distinction, and might or might not be a correct description of human cognition. That is, there might be kinds of attention happening at all levels of processing in human perception and cognition, not just at one dividing line between pre-attentive and post-attentive.

> Can you help me understand why Broadbent's filter theory
> of attention is incomplete? Thanks

Broadbent's filter theory asserted that pre-attentive processing was very low-level in terms of the type of information extracted, so that attention could be directed to low-level, "physical" aspects of stimuli such as pitch, location, etc., but not to high-level aspects such as meaning.

Subsequent data from Moray and Treisman et al. (p.54 of Reed) were problematic because they showed that attention could be diverted by the meaning of stimuli. So Broadbent's filter theory is incomplete.