1. . c, b, n, I, d, k, h, m, a, e or o

2. a b c d e j g n k l

3. a c j k j g n q l r

4. Lateral inhibition involves a line or sheet of neurons having inhibitory connections to the neurons on either side of them (in addition to their main signaling function). In the limulus eye this involves each unit (lens plus neuron system) having inhibitory connections with its neighbor (as well as it’s “primary mission” of passing excitatory information along.) Cells in its visual system (behind each little lens or facet) respond to the light illuminating them with excitation and also inhibit their neighboring cells, with less inhibition of cells further away and more for those that are near neighbors. This has the effect of enhancing contours or edges (places where the level of illumination changes) so that there is enhanced sensitivity to changes in illumination rather than absolute values.

5. Dulplex Theory of Vision. Cone based system (mainly but not exclusively) in center of eye that yields color experience, is very high acuity but requires high levels of light. It has (in center of fovea) one to one connection to ganglion cells.) Rod based system that does not yield color, mainly in periphery of eye, very sensitive to light so it can operate at low light levels. If the cone system was missing, you would have very little or no central vision, would not see the world in color and would not have much visual acuity. A missing rod system would mean that you would need high levels of light in order to see and would not have very little peripheral vision. Given that the cone based system is more sensitive to longer wavelengths and thus plays the major role in seeing the yellow car, as the light level declines at dusk or dawn, the cones stop working (not enough light) so the yellow car appears darker faster than a car whose reflected light is in the rod’s range of sensitivity (wavelengths that appear blue or green). Thus the green car will appear much brighter than the yellow when it gets darker because the rods still work and the cones don’t at the reduced light level.

6. The major result of M & S was that related words were faster in people’s responses to whether they were both words or not. This is because one word “primes” another (spreads activation to it). This is a large part of the basis for the Anderson ACT model of LTM which has words or conecpts/propositions connected by links to related words or concepts/propositions. ACT posits that our declarative memory consists of propositions stored in nodes connected to other propositions via links of some type over which activation can spread.

7. Sperling used 3 lines of four letters each, briefly flashed, and the whole report/partial report technique. With the whole report, subjects could report only an average of 4.5 letters, but with the partial report where you ask for a row of letters after they are "gone" you can show that virtually the whole image (10.5 letters av. Performance—3.5 out of the asked for row of 4) is stored. Delaying the "asking" for a row, progressivly reduces this performance until after a one second delay it is the same as the whole report performance level. He showed that we do have a short storage of the visual icon for at least half a second.

8. The basic idea of STM is that we have a short term store for a limited amount of “incoming” information (7+/- chunks that dieappear due to decay or interference within a short time. Working memory is the view that we have a small set of things that we can keep in mind that represent the activated part of LTM or the set of things we are thinking about at any given time—the current contents of consciousness is another way of viewing it. The focus of this is on activated LTM material, not on info encoded from sensory storage. The Baddeley view is that we have a verbal store of info that we can rehearse within a bit more than a second or so—as much as we can fit onto this articulatory loop we can keep in our short term store. There is also a visual-spatial sketchpad or space for spatial info. The focus of this view thus accommodates the dual code kinds of findings better and supplies an explanation for the importance of naming time on how much we can hold on the articulatory loop.

9. "Encoding specificity" refers to the finding that when we learn something we encode along with that learning, information about the context within which we learned it. Thus when we go to retrieve it, it is helpful to have much of that context present to help provide cues for retrieval. The experiment had people learn and retrieve in either a land-based or underwater in diving gear based setting. They did better when the retrieval context matched the learning context. Applying ACT to this, the implication is that the contextual information gets encoded along with the actual to be learned information. Thus if you are retrieving in the same context as when you learned the material, activation can spread from the context to the remembered material because they were learned together and thus linked in memory, and activation can spread over those links helping to activated the sought after information.

10. . heuristic, algorithm, start/goal/move operator/knowledge states/nodes/links/branchiness/length, 81, well-defined, ill-defined, set, functional fixedness, subgoals

11. a) Sensorimotor refers to the first couple years of life where the child learns to make sense of their sensory world (acquiring object constancy for ex.and use of internal representations) by the end of the period) and also achieve deliberate control over their motor responses. B) formal operations: 11 + years. Acquire ability to use logical and systematic thought in abstract problems. Use deductive and inductive thought processes and deal with multiple variables simultaneously.

c) Concrete operations age 7 to 11. Thinking becomes more systematic and logical. Success at conservation tasks (number, mass, volume, etc.) Decentration (no longer captured by single outstanding perceptual dimension, but able to classify objects in terms of more than one dimension at a time. d) habituation. Decreased response or loss of interest as a result of multiple presentations. Used in study of infant/child development, often to test whether they can discriminate new from old or different categories. E) segmentation. This refers to language acquisition whereby infants/toddlers have to learn word boundaries (segment the speech stream which is more or less continuous, into separate words. F) A gene that is tied to serious language deficiencies if absent. G) the developmental concept that others have points of view, experiences that are like ours but not simply ours….that other people have minds.

12. a) In the Kosslyn task there was a linear relationship between time to “scan” from one location to another, demonstrating that the representation was spatial in nature. If that wasn’t the case, the linear relationship would not have held. b) In Shepard and Metler, the result was that the amount of time to answer whether the two block figures could be matched or not was linearly related to the angle of rotation, suggesting something akin to mental rotation. If there was one code this result (of an analog to actual rotation) would not have been obtained. c) in Brooks’ experiment, a spatial output task (pointing at Y or N for yes or no) interfered with a spatial (scanning the block letter and judging the vertices) but not a verbal task (part of speech judgements about a remembered sentence). A verbal response task did not interfere with the spatial block letter task but did with the verbal task. This gave evidence of a visual-spatial code in memory or thinking. If such a code didn’t exist, the above results would not have been (the crossed pattern of interference in the Brooks task).

13. d b c e b d c b c c

e.c. You might use hill-climbing in which you try to get closer to the apparent goal on each move. However you would soon find that you can't do that--after moving the small disk to the goal peg (the only first move that seems to get you closer) you can't put any other disk on top of it and are blocked! Means-ends-analysis allows you to set subgoals where you can adopt the subgoal of getting a disk out of the way to allow another to be moved. It can solve the problem (via decomposition--breaking up the task into goals and subgoals). Other heuristics are also possible, such as working backwards (starting with the disks on the goal peg and exploring reversing moves to move them back to the start.