What follows is a scoring guide used to grade the Midterm Exam.
On the initial scoring of the exam, performance was perhaps a little low, with an average score of 32.26 (SD = 9.37), corresponding to 65%. Psychometric analysis yielded a pretty decent reliability (Cronbach's alpha) of .81.
No item had a low item-to-total
correlation, so there were no
"bad" items by that standard.
To identify "difficult" items, I examined the percentage scores for each item (most items were worth 4 points, but some items were worth only 1 point). The average item score, computed as a percentage of the points available for that item, was 64% (SD= 14%).
No item appeared to be an "outlier", with an average percentage score was more than 2 SDs below the average of all items. However, two items came close: #s 9 and 10. So I eliminated these "iffy" items by giving everyone full credit, raising the mean exam score further to 37.12 (SD = 7.77).
That's still a little low (frankly). My intention is to keep exam difficulty relatively constant from year to year. In order to bring the mean score up to those of previous years' midterm exams, I added 5 points to everyone's score, truncating scores at 50. This resulted in a final mean exam score of 41.84 (SD = 7.45). The median score was 44. The figure at left shows the distribution of scores.
Exams will be returned in discussion sections, the
week of October 26. If you think a clerical
error has been made in computing your grade, contact
Prof. Kihlstrom. If you think a substantive error has
been made in grading a test question, give your GSI a
written paragraph detailing why your answer is better
than the one given in the scoring guide.
The deadline for requesting a
rescore is Monday, November 3, 2014 at 5 PM.
Answer each the following 14 questions. Each question is worth 1-4 points, so that the exam totals 50 points.
Do not provide long-winded answers. You have about three minutes, on average, for each question, and we are grading accordingly. Use only the space provided for your answer. Each question requires no more than five (5) sentences. Go through the entire exam quickly to identify the questions you can answer easily. Do them, then turn to the others. In either case, get right to the point.
Read through the entire exam first. Then go back and answer the questions you find easiest. Then turn to the harder ones.
Write your name at the top of every page, so that the exams can be separated for grading.
Write your answers in ink. Answers written in pencil will not be eligible for regarding.
And write legibly, or we won’t be able to appreciate how wonderful your answers are. Complete sentences are always nice, but they’re not absolutely necessary.
The scoring guide used to grade the exam will be posted to the course website as soon as possible after the exam.
Exam grades will be posted on the course website as soon as possible after October 27.
Exams will be returned in sections on October 29-30.
Also, please indicate your Discussion Section # (or time, or GSI) here:_____________
No Office Hours Today (Wednesday).
Sections will not meet this week.
Section 1: Introduction.
1. When he talks about “conscious shyness”, how does the philosopher Owen Flanagan distinguish between “conscious inessentialism” and “epiphenomenalism”? (1 point) [Lecture]
M score = 0.68 points; item-to-total rpb = .55. Conscious inessentialism is the idea that consciousness is not necessary for the performance of many, if not most or even all, mental functions. But unlike the epiphenomenalist suspicion, it still allows consciousness to play a causal role in behavior.
Part 2: Introspection
2. List and
briefly define the “Five Characters of Consciousness”,
according to William James.
(1 point for 1; 2 points for 2; 3 points for 3 or 4; 4
points for all 5) [Lecture;
Revonsuo, Chapter 2]
=2.88, rpb = .31.Here are the "Five
Characters", in order:
1. Personal Subjectivity: Mental states ae experienced as belonging to, or owned by, the individual.
2. Constant Change: No two conscious states are ever identical.
3. Continuity: Consciousness occurs in a continuous stream, from awakening to falling asleep.
4. Intentionality: Conscious mental states are always about something.
5. Selective Attention: We are only aware of a small set of possible objects and events in the environment.
3. What are “qualia”? Orange is one of Newton’s seven “primary colors”, but it is not one of the elementary qualia of visual experience. Why not? (1 point for the definition; 1 point each for correct reasons, up to 4 points) [Lecture; Revonsuo, Chapter 3]
2.40, .36. Qualia are the elements of conscious experience. “Orange” doesn’t meet at least three of Dennett’s four properties of qualia. (1) It’s not ineffable, because you can describe orange as what an orange looks like. (2) And it’s not unanalyzable because you can analyze it into red and yellow. (3) It’s not private, because one person can say to another, ‘It doesn’t look like an orange to me”. And it doesn’t match many of Cutting’s properties either. (1) Orange is itself the result of a physical mixture. (2) It doesn’t have a dedicated neural pathway, the way red, green, yellow, and blue do. (3) In language, the color term orange appears only in languages with a rich vocabulary of color. (4) The term orange refers to an object, an orange, in a way that terms such as red and green do not..
4. What do some
theorists distrust introspective self-reports? (Any 4 reasons, for
up to 4 points) [Revonsuo,
2.99, .41. Here are a bunch of valid reasons, others are possible.
1. They’re private, and can’t be publicly confirmed.
2. Because all introspection is retrospection, subjects may forget what an experience is like by the time they report on it.
3. Or they may reconstruct the experience, based on their expectations and implicit theories, and thus distort it.
4. They may fill in the missing gaps in their experience by confabulation.
5. It may be difficult to verbalize what an experience is like.
6. The very act of introspection may alter the experience.
7. Subjects may respond to the “demand characteristics” of the experimental situation and give the experimenter what he or she want to hear.
8. Subjects may censor their experiences, and not report them accurately.
9. Some important aspects of mental life may not be accessible to consciousness at all.
Part 3: Mind and Body
5. Where does the “Cartesian impasse” arise, and how does it relate to the “explanatory gap”? (4 points) [Lectures and Revonsuo Chapter 1]
3.64, .37. The Cartesian impasse refers to the problem of explaining how mind and body, which seem (to Descartes, and most of the rest of us) so different, can interact with each other. Bodily activities cause mental states, as when the visual system gives rise to the conscious experience of seeing. And mental states give rise to bodily states, as when the belief that one is taking a potent analgesic relieves pain. The explanatory gap is another metaphor for the same problem: We can have an explanation for how the brain works (which is what neuroscience does), and we can have an explanation for how the mind works (which is what psychology and cognitive science do), without being able to put those two explanations together to explain how we get conscious mental states out of brain-processes.
6. What does Dennett mean by the “Cartesian Theater” model of consciousness, and why does he think it’s wrong? What is his “Multiple Drafts” model, and why does he think it is preferable? (2 points for explicating the Cartesian Theater; 2 Points for the Multiple Drafts model, for a total of 4 points) [Searle, Chapter 5; Revonsuo chapter 10]
.47. The Cartesian Theatre is the idea that there
is a place in the brain where “it all comes together”, and we
experience our mental states much the way a person (the
homunculus?) might view a play or a movie. Dennett believes
it’s wrong that we have subjective states in the first place,
so there’s no need to postulate a place where they occur. Instead, his
Multiple-Drafts model postulates a series of informational
states standing between stimulus inputs and response outputs –
like the “hidden levels” of a connectionist model. We might be aware of
any of these, or none, but our subjective awareness is purely
fact that we can describe, one of these intermediate states,
through verbal or nonverbal behavior, makes no difference to
the operation of the “virtual machine” that is the brain.
7. Why does Chalmers’s theory of consciousness count as a version of dualism? What kind of dualism is it? What else is unusual about it? (4 points)
2.62, .53. Chalmers posits that experiential states are a fundamental feature of the universe, but not part of the material world - -that makes his theory dualistic. But it’s not an interactionist form of dualism, because he can imagine a “zombie world” that is materially identical to ours but which lacks consciousness (in this way, he seems to embrace conscious inessentialism). Moreover, he believes that experiential states have no causal impact on the world – which makes his theory a version of epiphenomenalism. For good measure, he attributes experiential states to any material system that represents information, including thermostats and solar systems – thus embracing panpsychism as well. To put it bluntly, Chalmers believes that consciousness is (almost) everywhere, but has nothing to do with how the real world operates. [Searle, Chapter 6; Revonsuo Chapter 10].
Part 4: Attention and Automaticity
8. What are the implications of early- and late-selection theories of attention for preconscious processing, and how do capacity theory attempt to resolve the debate? (4 points) [Lectures]
2.72, .59. Early-selection theories posit a filter (or attenuator) which selects information for further processing based on an analysis of its physical properties – such as the spatial location of its source. Therefore, preconscious processing is limited to physical analysis and the creation of “perception”-based representations. Late-selection theories assert that all available information is fully processed, including semantic analysis, so that preconscious processing can include analyses of meaning as well as of physical features. Capacity theories permit even complex semantic analyses to be performed preconsciously, so long as the process has been automatized. While some physical analyses may be innately automatic, even complex semantic analyses can be automatized, provided that the subject has had sufficient practice performing them. .
9. Without going into the mathematics, explain how Jacoby’s “Process-Dissociation Procedure” estimates the contribution of automatic and controlled processes to task performance? What are the typical results of such comparisons? (4 points) [Lectures]
1.52, .59. Not a bad item, but its average was low enough to be "iffy". Jacoby’s PDP employs the Method of Opposition to pit automatic and controlled processes against each other. Under the “Inclusion” condition, task performance is influenced by both controlled and automatic processes. Under the “Exclusion” condition, performance is influenced solely by automatic processes. Therefore, what is, essentially a subtractive logic yields estimates of the two operating independently. PDP comparisons rarely show that automatic processes dominate controlled processes, except under specialized conditions that effectively limit the scope of cognitive control.
10. What are the differences between the “first” and “second” version of Crick and Koch’s neurobiological theory of consciousness? (4 points) [Revonsuo, Chapter 11]
1.64, .46. Not bad, but also "iffy". Crick and Koch originally proposed that consciousness is generated by the synchronous firing of ensembles of neurons (what they called essential nodes), each representing some aspect (otherwise known as a quale) of sensory perception – say, the color red and the shape of a T). When these nodes fire together, in a coalition of essential nodes, at a rate of 40 cycles per second (in what is known as the gamma band), the subject is consciously aware of the stimulus – say, of a red T. But then they realized that synchronized firing would produce only the integration of individual features into a unified percept – not necessarily a conscious one. Later, they argues that different coalitions of essential nodes competed for association with other neural ensembles elsewhere in the cortex, and also in the thalamus, and that the “winner” of this competition gets represented in consciousness
Part 5: The Explicit and the Implicit
11. How is the distinction between repetition priming and semantic priming relevant to the debate over the scope of unconscious (“subliminal”) perception? (4 points) [Lecture]
3.01, .49. All theories of attention, including early-selection theory, agree that the physical features of a stimulus can be processed preattentively. So, if subliminal perception occurs at all, it ought to be manifested in repetition priming, which requires nothing more than the construction of a perception-based representation of the prime. The real question is whether we can get more than repetition priming from a “subliminal” stimulus. And the answer is yes: Marcel got evidence for priming based on denotative meaning (as in doctor-nurse or the subliminal Stroop effect)), while Greenwald got evidence for priming based on connotative meaning (as in friend-wins). So the fact that we get semantic priming at all, even if it’s limited in terms of duration or analytic scope, would seem to be inconsistent with early-selection theories.
12. Distinguish between a “single” dissociation and a “double” dissociation in the context of explicit and implicit memory. Why do we need the latter if we’ve got the former? (4 points) [Revonsuo Chapter 5]
2.10, .29. Implicit memory can be dissociated from explicit memory, as when priming effects occur in the absence of conscious recall or recollection. In a single dissociation, a single experimental variables (like hippocampal damage or level of processing) affects performance on one task (such as recall or recognition) but not the other (such as priming). But in order to rule out confounds, such as task differences in difficulty, it’s good to be able to show that there is some other experimental variable which has the opposite effect – say, affecting priming without affecting recall or recognition. Even better evidence comes from a crossover dissociation, in which a single variable has opposite effects on the two tasks.
13. The Implicit Association Test appear to show that even “unprejudiced” white subjects unconsciously harbor activate negative stereotypes and attitudes toward racial and ethnic out-groups. Without going into details on stimulus-response compatibility, explain how the IAT do this, and indicate at least one problem with this inference. (2 points for an explication of the IAT, and 1 point for each potential problem, for a total of 4 points) [Lecture]
2.94, .32. The uses response latencies in a pair of classification tasks to measure the strength of association between, say, an ethnic group (like Swedes) and the concept of “good” (or pleasantness or social desirability) and another ethnic group (like Norwegians) and the concept of “bad” (or unpleasantness or social undesirability). As for interpretive difficulties, any two of the following will do.
Section 1: Introduction (Again).
14. When he talks about “conscious shyness”, what does the philosopher Owen Flanagan mean by “positivistic reserve”? (1 point)
0.57, .31. Positivistic reserve is a preference for objective behavioral data (including physiological data), as opposed to self-reports.
This page last revised 10/28/2014.