Basic Learning Processes Final
Retrospective memory
'looking back' memory If you are a rat on the radial maze, you could solve the problem by remembering where you have been. But problems become more difficult as the memory load increases.
Prospective memory
'looking forward' memory If you are a rat on the radial maze, you could solve the problem by remembering where you need to go. Problems get easier as you go along.
Lexigrams are used in ______.
...language acquisition in chimps.
Chimps have been observed using ______.
...make-shift spears to hunt Bush Babies.
Honig proposed _______.
...that memory takes two forms: reference and working memory.
What are some causes of hippocampal damage?
1. Experimental lesions: Electrolytic or chemical lesions. 2. Toxic environmental chemicals: Heavy metals such as lead, tin, mercury. 3. Accidents: Carbon monoxide poisoning or asphyxiation. 4. Diseases: Hippocampus is hit early and hard in Alzheimer's disease.
Retrograde amnesia
A gradient of memory loss going back in time from the occurrence of a major injury or physiological disturbance. Amnesia is greatest for events that took place closest to the time of injury and less for events experienced earlier. Can be caused by ECS (electroconvulsive shock), anesthesia, hypothermia, hyperthermia (e.g., fever), drugs.
Infantile amnesia
A lack of explicit memory for events that occurred before the age of roughly 3.5 years. While people are unable to recall memories from his part, learning and memory do still occur.
Cognitive Map
A mental representation of the layout of one's environment.
The Scalar Expectancy Theory
A model of timing. What might be the details of a mechanism that permits organisms to respond on the basis of temporal information? Time is not a physical reality. It is a human invention that helps us characterize certain aspects of our environment. We cannot see a time interval; one can only see the events that start and end the interval. Given that time itself is a conceptual abstraction, models of timing also tend to be fairly abstract.
Delayed matching-to-sample procedure
A procedure in which participants are reinforced for responding to a test stimulus that is the same as a sample stimulus that was presented some time earlier.
What is a radial maze?
A radial maze is a maze with several pathways (8 to 17) radiating outward from its center. The maze was introduced by Olton & Samuelson (1976). Typically the 'arms' of the maze are 'baited' (food reward) and the rats are allowed to freely decide which route to take. Rats chose randomly and remembered where they had already been to avoid returning to an arm without food. Rats developed a cognitive map of the maze. Later studies showed that the hippocampus was important in memory, such that the rat did not return to a previously visited arm.
Beatty & Shavalia (1980)
Active control of working memory. Found that working memory lasts at least 8-12 hours.
Pepperberg and Alex the ______
African Grey Parrot Alex was trained daily, almost constantly, using the model-rival technique. Alex observed an model-rival pair and joined in the social interaction. Alex was taught to use language to: - Talk to humans - Ask for things - Count up to 6 things
Dashiell (1930)
Alley maze Memory flexibility. Rats are not 'little automatons.' A single reward (food box) location was available but several routes, each of equal distance to reward, could be traveled to retrieve reward. The rats learned to run in the direction of the food box without fixating on any particular pathway, yet without making any errors. In the literature, this is known as 'goal-direction orientation' - the rats were able to orient themselves towards the goal/reward irrespective of a particular/single pathway.
Anthropomorphism
Attributing human characteristics to an animal.
Roberts (1981)
Blackout procedure Roberts introduced the 'peak procedure' to study timing in rodents. Example: https://www.youtube.com/watch?v=eT7uqR1498k&spfreload=10 (notice how the animal begins responding a lot when it's close to the time of reinforcement). [Quick summary: Rats recognize the point in time when reinforcement is coming (peak procedure). So even when a 'blackout' occurs, they can recognize that time. So if the peak occurs at 40 seconds, and along the way there's a 10-second 'blackout' - they will delay their gradually increasing responding accordingly.] During a blackout procedure, a tone signaling a 40-sec interval started a trial. After 20 sec of this interval, the lights were turned off for 10 seconds before being re-illuminated. Because of the blackout period, the rats responded 50 seconds after the the start of the trial rather than 40 seconds later. These results suggest that a rat can stop and restart its internal clock based on an estimation of the illuminated interval.
Part 2: Briggs, Riccio (2007); Double-bump hypothesis
Briggs and Riccio (2007) repeated the experiment I described in Part 1, but this time they added three groups whose body temperatures were re-cooled for the memory test. The 're-cooled' temperatures matched the temperatures at which the rats experienced hypothermia. Refer to the graph: These results show that amnesia caused by hypothermia can be reversed by returning the subjects to the body temperature they had right after the extinction procedure. Thus, hypothermia did not disrupt consolidation of the memory of extinction, but made that memory difficult to retrieve when the rats were at normal temperature. The implication of this type of experiment is that one's memory for the details of a car accident could be restored by reenacting many of the elements of the accident. This supported the retrieval failure hypothesis.
Retrieval Failure Hypothesis
By David Riccio. The status of memories following experimentally induced amnesias: Gone, but not forgotten. A deficit in recovering information from a memory store.
What animals are good at language? How?
Chimps, like Kanzi the bonobo, are very good at using arbitrary stimuli as symbols and can use simple grammar. Dolphins can understand the same meaning from sentences with different word order.
Terrace (1987)
Chunking in pigeon list learning Chunking allows for faster acquisition of list learning. Chunking: Grouping information in series to increase memory retention. Creating a segregated list in your mind. Goal was to see if the pigeons had the ability to detect the categories and improve their memory due to 'chunking.' Sequence learning: Pigeons tasked with pecking objects on a screen in the correct order. If it makes an incorrect response, then the screen goes blank and they have to start over again. Begins with 2 objects (max is 5). If it pecks correctly, reinforced with grain. More items are added onto the list over time. Starting with low number of items to train with, and then over time as the pigeon learns - you can add more elements to the sequence. This is known as chaining. Terrace believed that the pigeons would learn faster and more effectively if 'like was grouped with like.' Hence, a segregated sequence would be more effective than a mixed sequence.
Brannon and Terrace (1998)
Counting in rhesus monkeys [Quick summary: Brannon and Terrace showed that monkeys can count to 9 and perhaps beyond.] Two monkeys were first trained to respond to exemplars (something typical or representative of a class of things) of the numerosities (something that can be counted) 1 to 4 in an ascending numerical order (1 → 2 → 3 → 4). As a control for non-numerical cues, exemplars were varied with respect to size, shape, and color. The monkeys were later tested, without reward, on their ability to order stimulus pairs composed of the novel numerosities 5 to 9. Both monkeys responded in an ascending order to the novel numerosities. These results show that rhesus monkeys represent the numerosities 1 to 9 on an ordinal scale.
Tool use in Corvids
Crows bent paper clips to lift a food reward out of its container.
Milmine, Watanabe, & Colombo (2008)
Directed Forgetting Direct forgetting is forgetting that occurs because of a stimulus (a 'forget cue') that indicates that working memory will not be tested on that trial. Directed forgetting is an example of the stimulus control of memory. Demonstrations of directed forgetting are important because they provide evidence that memory is an ACTIVE process that can be brought under stimulus control. Procedure: The experimental chamber had three pecking keys arranged in a row. The center key was used to display the sample stimulus (a red or white light), and the two side keys were used during tests of memory, which involved a choice between the sample and the alternate color. Five different types of trials could take place. Each trial started with a sample stimulus presented on the center key. This was followed by different auditory cues that served as the R-cue (remember), F-cue (forget), or free-reward cue. Probe trials evaluated the impact of the F-cue and free-reward cue on memory. Results: Most accurate matching performance occurred on R-cue trials. As expected, the pigeons did poorly on the F-cue trials, indicating that the F-cue disrupted memory. Notice that because the forget instruction was provided after the sample stimulus, one cannot argue that the F-cue disrupted attention to the sample. Rather, the F-cue altered memory mechanisms (rehearsal) during the delay interval.
Herman and ______
Dolphin language comprehension. Novel sentence grammars and tests.
Retrospective + prospective memory = ?
Easier time remembering overall. Both rats and human subjects can do this. This is strong evidence for the active control of memory in animals.
Duncan (1949)
Electroconvulsive shock (ECS) and memory [Quick summary: Memory improved the longer the time elapsed between avoidance training and ECS. Better memory consolidation with time. Memory takes some time to consolidate after a learning experience.] Researchers in the 40s and 50s used ECS to disrupt consolidation of memory active-avoidance conditioning trials and other learning situations. Recall that an active avoidance trial tasks the subject with making an avoidance response to avoid the aversive stimulus. In general, ECS was found to produce retrograde amnesia - an inability to recall events that precede a traumatic event, resulting in a failure to show the learned avoidance behavior in later trials. In Duncan's experiment, rats learned to run from a dark chamber with a grid floor (through which shocks could be passed) into a lighted safe compartment with a wire mesh floor. After 10 seconds in the safe chamber, the rats were removed and then given ECS a varying intervals after the learning trial, or no ECS if they were control animals. Consolidation seemed to occur relatively quickly, as rats given ECS after longer intervals (1, 4, and 14 hours) did not differ from a control group that had the training without receiving ECS.
Clayton, Bussey, and Dickinson (2003)
Episodic memory in Jays Episodic memory is memory for a specific event or episode, as contrasted with memory for general facts or ways of doing things. The Western scrub jay is an ideal species for studying questions related to episodic memory because it caches both perishable and nonperishable food, engages in caching behavior all year round, and readily performs these activities in the laboratory. Western scrub jays prefer to eat worms over peanuts. However, worms are perishable and deteriorate if they are stored a long time. Clayton and Dickinson first gave jays practice trials in which they were allowed to store worms and peanuts in the compartments of an ice cube tray. A different tray was used for each type of food. The trays were made distinctive by placing different objects around them. To permit hiding the foods, the compartments of each ice tray were filled with sand. Each trial consisted of two storage or caching episodes (one for peanuts and the other for worms, in a counterbalanced order). A recovery period was then conducted four hours or 124 hours later with both food trays available. On training trials with a four-hour retention interval, neither food deteriorated by the time the recovery or choice test occurred. In contrast, on trials with the 124-hour retention interval, the worms were in pretty bad shape by the time of recovery. As training progressed, the birds learned to select the worms during the recovery period if recovery was scheduled four hours after caching. This is because the worms would still be intact. If recovery occurred 124 hours after caching, the birds selected the peanuts instead. However, this behavior could have been cued by the sight or smell of the pea- nut and worm caches during the recovery period. To prevent responding on the basis of visual or olfactory cues, test trials were conducted at the end of the experiment during which fresh sand was put in the trays and all food was removed. As expected, these birds searched more in the worm tray than the peanut tray if the choice test occurred four hours after caching of the worms. In contrast, they searched more in the peanut tray than in the worm tray if the worms had been stored 124 hours earlier.
Clayton, Bussey, Dickinson (2003)
Episodic memory in Scrub Jays. 2 groups: deteriorate and replenish. In the deteriorate group, the birds knew that their worms wouldn't last forever. So after 124 hours, they preferred the peanuts. In the replenish group, where worms were replenished by experimenters, the birds went for the worms after 124 hours.
Sargisson & White (2001)
Evidence AGAINST Trace Decay In Sargisson & White's study, different delay intervals were introduced to see if better memory could be trained by using longer delay intervals from the BEGINNING of training. This is in contrast to a typical delayed-matching-to-sample-task where the subjects are initially trained without a delay. Procedure: For one group of pigeons, the delay between the sample and choice stimuli was always 2 seconds during training. For other groups, this delay was 4 seconds or 6 seconds. For comparison, a control group was trained with the typical procedure of presenting the choice alternatives immediately after the sample (no delay). Training was continued until each group performed correctly on at least 80% of the trials. All of the birds were then tested with delays ranging from 0 to 10 seconds, to determine their forgetting functions. Results: The control group that had been trained with a 0-second sample-choice delay showed the standard forgetting function. Their rate of errors increased as the delay between the sample and choice stimuli was increased from 0 to 10 seconds. In contrast, no such decline was evident in pigeons, which had been trained with a 6-second delay between the sample and choice stimuli. These birds performed equally well at all test delays. For all groups - the most accurate performance occurred when the delay used in the test was the same as the delay that they received during training. These results clearly show that forgetting functions do not directly reflect the decay or fading of memory for the sample stimulus as a function of time. Rather, test performance depends on the similarity between the conditions of testing and the conditions of training. So why did Grant come to different conclusions? The common finding that memory gets worse with the passage of time may simply reflect the fact that participants do not have practice with longer delay intervals.
Studies by Mark Rilling and Hank Davis: Directed Forgetting
Forgetting that occurs because of a stimulus (a forget cue) that indicates that working memory will not be tested on that trial. Directed forgetting is an example of the stimulus control of memory.
What kinds of evidence would help decide between the memory-consolidation and retrieval-failure interpretations?
If information is lost because of a failure of consolidation, **it cannot ever be recovered.** By contrast, the retrieval-failure view assumes that amnesia can be reversed if the proper procedure is found to reactivate the memory. Thus, to decide between the alternatives, we have to find techniques that can reverse the effects of amnesic agents. Contextual cues of acquisition and unpaired presentations of the US are good potential candidates to reactivate memories.
What is the contemporary view of memory storage and recall?
In the contemporary view, memory is either in an active or inactive state. Active memories can always be consolidated for transfer to the inactive state, whether the information is in the active state because of original learning or because of the activation of an old memory.
The bonobo chimpanzee most famous for learning to use lexigrams is ______.
Kanzi
Jane Goodall (1960)
Monkeys using tools; hooded monkeys using plastic tubes to get yogurt.
Baran et al.
Monkeys were tasked with a match-to-sample experiment, using a touchscreen.
How do rats know if they've been down a runway?
Odor. Rats have excellent sense of smell. Additionally, the maze was made of lumber so it could have had environmental cues. For example, splinters or scratches.
Grant (1976)
Passive Trace Decay (Quick summary: Pigeons performed better [higher percentage of correct responses] when the sample duration was longer AND when the delay between presentation of the sample stimuli and the opportunity to make a choice was short. Notice that the rate of decay was constant across the 2 dimensions [sample duration and delay].) Grant tested pigeons in a Skinner box that had three pecking keys in a row on one wall above a food hopper. The stimuli were colors (red, green, blue, and yellow) which could be projected on the pecking keys. At the start of each trial, the center key was illuminated with a white light, to signal the start of the trial. The pigeon was required to peck the start cue to make sure it was facing the response keys. After the pigeon pecked the start cue, the sample color for that trial was presented on the center key for 1, 4, 8, or 14 seconds. This was followed by delay intervals of 0, 20, 40, or 60 seconds, after which the two side keys were illuminated, one with the sample-matching color and the other with an alternative color. After the bird made its choice, all the keys were turned off for a 2-minute intertrial interval. Results: Accuracy in the delayed-matching-to-sample procedure decreased as a function of the delay interval and increased as a function of the duration of exposure to the sample stimulus. The 'trace decay hypothesis' explains these results. This hypothesis assumes that presentation of a stimulus produces changes in the nervous system that gradually dissipate, or decay, after the stimulus is turned off. The initial strength of the stimulus 'trace' is assumed to reflect the physical energy of the stimulus. Thus, longer or more intense stimuli are presumed to produce stronger stimulus traces. Increasing the delay interval in the matching-to-sample procedure reduces the accuracy of performance, because the trace of the sample stimulus is weaker after longer delays. By contrast, increasing the duration of exposure to the sample improves performance, presumably because longer stimulus exposures establish stronger stimulus traces.
Reference Memory
Permanent memory. Long-term retention of background information necessary for successful use of incoming and recently acquired information. (Compare with working memory.)
Baran et al. (2012)
Prospective coding in matching-to-sample The matching-to-sample procedure can be solved with either retrospective or prospective memory. Beran et al. used a modified procedure to encourage prospective coding. The experiment was conducted with rhesus and capuchin monkeys. The monkeys previously served in numerous experiments that required them to manipulate a joystick to move a cursor on a computer screen - so by the time of this experiment, they had become very efficient at using the joystick. At the outset of each trial in the present experiment, a sample stimulus was presented in the center of the computer screen, with four possible choice cues arranged around the sample. The stimuli were selected from a set of 500 different images. If the monkeys selected the choice alternative that matched the center sample, they received a bit of food. The sample image varied from trial to trial, and where the sample appeared among the choice alternatives at the corners of the display also varied from trial to trial. To force the monkeys to plan their future actions, the choice alternative were masked as soon as they started to move the cursor. Thus, to get the cursor to the correct corner of the display, the monkeys had to remember where they intended to move the cursor. Another challenge was that the monkeys had to first move the cursor to the sample stimulus and then to the correct choice alternative. This was difficult but eventually 4 out of 5 capuchins and 5 of the rhesus monkeys became very skillful in performing the task.
Olton and Samuelson (1976)
Radial maze procedure [Quick summary: Rats do not require much training to perform efficiently in the radial maze. The radial maze task takes advantage of foraging tactics that rats acquired through their evolutionary history. Rats do not use rules; they choose randomly and simply remember where they have been to avoid returning there. They develop 'cognitive maps' of the maze. Later research showed that the hippocampus of the rat brain was critical for remembering where the rat had been. Damage to the hippocampus impairs their capacity to remember where they've been.] A radial arm maze typically has eight arms radiating from a central choice area, and there is a food cup at the end of each arm (Olton & Samuel- son, 1976). Before the start of each trial, a pellet of food is placed in each food cup. The rat is then placed in the center of the maze and allowed to go from one arm to another and pick up all the food. Once a food pellet has been consumed, that arm of the maze remains empty for the rest of the trial. Given this contingency, the most efficient way for a rat to get all eight pellets is to enter only those arms of the maze that it had not yet visited. That is, in fact, what rats do. Entering an arm that had not been visited previously was considered to be a correct choice. During the first five test runs after familiarization with the maze, the rats made a mean of nearly seven correct choices during each test. Rats live in burrows, and outings to find food are followed by return to the home burrow. While out foraging, they follow preexisting trails and move about without returning to recently visited places. Their tendency to avoid recently visited places is so strong that they don't return to recently visited arms in a maze whether or not the maze arms are baited with food at the start of the trial. These results suggest that radial maze performance has deep evolutionary roots. ***How are rats able to do this? Rats appear to use distinctive features of the environment, such as a window, door, corner of the room, or poster on the wall as landmarks, and locate maze arms relative to these landmarks. Movement of landmarks relative to the maze causes the rats to treat the maze arms as being in new locations. Thus, under ordinary circumstances, spatial location is identified relative to distal room cues, not to local stimuli inside the maze. (Similar spatial cues are involved in guiding successful performance in the Morris water maze.)
Describe Dashiell's alley maze experiment (1930)
Rats placed in a maze that was symmetrical with equidistant routes. Each time, the rats chose a different path. However, all of their choices were 'good' routes. This was ultimately evidence of flexible use of memory. Animals prefer variability. Once they made a cognitive map of their environment, they were able to navigate using their spatial memory.
Part 1: Briggs, Riccio (2007); Retrograde amnesia effect on extinction
Rats were tested for the effects of retrograde amnesia on extinction. [Quick summary: Rats were shocked in a black box. There were several experimental groups. Some rats received hypothermia instantaneously, 30 minutes, and 60 minutes after extinction. Both the Hypo-0 and Hypo-30 minute cohorts showed longer latencies. However, the 60 minute cohort showed an identical latency to the solely-extinction group. This proves effect of retrograde amnesia. Later, the experimenters tested the effect of body temp. the next day and found that the colder the rats were - the shorter their latency was.] Procedure: Laboratory rats were conditioned in a shuttle box that had a white and a black compartment separated by a door. Conditioning was accomplished in a single trial. The rats were placed in the white compartment and the door to the back compartment was then opened. As soon as the rats walked into the black compartment the door behind them closed and they got two inescapable shocks. This single-punishment episode made the rats reluctant to enter the black compartment again, and that was used as the measure of conditioning. Extinction was conducted the day after conditioning and consisted of putting the rats in the black compartment without shock for 12 minutes. Following the extinction procedure, some of the rats were immersed in cold water to substantially reduce their body temperature. This hypothermia treatment was provided either immediately after the extinction treatment (when it would disrupt consolidation of the extinction experience), or 30 or 60 minutes after extinction. The next day the rats were put back into the white compartment to see how long they would take to enter the black compartment (which had been paired with shock). Results: (Refer to picture) Notice that rats in the Hypo-0 group behaved as if they never got extinction. They took more than 500 seconds to enter the black compartment. Similar results were obtained with the rats in the Hypo-30 group. Thus, these two groups showed retrograde amnesia for the extinction treatment. However, delaying the hypothermia 60 minutes after extinction did not produce the amnesic effect. Rats in the Hypo-60 group responded like those in the normal extinction group, entering the black compartment in about 200 seconds.
Olton found, in contrast to Beatty & Shavalia, that _______.
Rats were virtually errorless when tested twice (8 choices each; which was also the maximum amount of choices they could make in the maze) only 15 min apart. Olton claimed that they must have actively "reset" working memory to erase it. This is evidence of 'active control of working memory.' Why would this be useful? After the rats make their 8 choices in the maze, they are taken out for 15 minutes. When they are put back in, they are once again tasked with making 8 choices in the maze. If they can't reset their working memory, then that earlier memory (of the choices they made) would INTERFERE with whatever choices they make in the present. So it's useful for them to be able to 'reset' their working memory. This is why rats make different choices when you put them into the maze (providing they had already completed the maze previously). The Beatty & Shavalia study had a different experimental design. In that study, the rats were allowed to make the first 4 choices and then they were removed from the maze for varying amounts of time. The experimenters were testing how long their working memory lasted because the rats hadn't completed the maze (since they were only allowed to make the first 4 choices).
Autonoetic consciousness
Reliving the past experience.
Kesner and DeSpain (1988)
Retrospective versus prospective coding Example of retrospective coding: - Animals keep in mind where they have ALREADY visited. Example of prospective coding: - Animals keep in mind which maze arms they have YET to visit. [Quick summary: Both rats and college students can use both retrospective and prospective coding effectively to accomplish a memory task. Memory load for retrospective coding is low in the beginning. Then as the rat visits more places, memory load increases and retrospective coding loses it's efficacy. Now, the rat will switch to prospective coding because remembering where it has NOT been yet will be easier at this point. So memory load for prospective coding decreases as the rat is near the end of the task. The results for rats and human subjects were nearly identical. Refer to the graph. This suggest that memory performance is a function of coding strategies and that coding strategies may vary as a function of task demands. Participants switch from one coding strategy to another so as to reduce memory load and thereby improve response accuracy.] This is strong evidence for active control of memory in animals. Kesner and DeSpain compared the coding strategies of rats and college students in spatial memory tasks. If individuals switch from retrospection to prospection in the course of remembering a series of places, memory load should first increase and then decrease. Memory load was estimated from the rate of errors the participants made on a test that was conducted after the participants had visited different numbers of places. Rats: The rats in Kesner and DeSpain's study were first trained to forage for food on a 12-arm radial maze in the standard manner. Once they had become proficient at obtaining food by going to each maze arm, a series of test trials were conducted. On each test trial, the rats were allowed to make a certain number of arm entries. They were then removed from the maze for 15 minutes. At the end of the delay, they were returned to the maze and allowed to enter one of two alleys selected by the experimenter. One was an alley they had entered earlier; the other was a previously unchosen alley. Selecting the new alley was judged to be the correct response. As the number of visited locations before the test increased from two to eight arms of the maze, the error rate increased. This finding is consistent with the hypothesis that the rats were using a retrospective coding strategy during the first eight arm entries. Interestingly, however, when the rats were tested after having entered 10 arms, they made fewer errors. This improvement in performance towards the end of the series suggests that the animals switched to a prospective coding strategy. (Refer to graph) College students: The college students in the study were presented with a grid containing 16 squares (corresponding to 16 places in a maze). During the course of a trial, the symbol X traveled from one square to another in an irregular order, simulating movement from one place to another in a maze. After the X had been presented at various locations, a delay of five seconds was introduced, followed by presentation of two test locations. One test location was a place where the X had been; the other was a new square. The participants had to identify which was the new square. The results were strikingly similar to the pattern of errors obtained with the rats. The error rate initially increased, consistent with a retrospective coding strategy. After the target stimulus had been in eight places, however, the error rate decreased, consistent with a prospective coding strategy.
Riccio's 'double-bump' hypothesis
Riccio proposed that if the learning context is reinstated, the inaccessible memory might be recovered. Evidence: 1. Re-exposure to the amnestic agent 2. Concussion studies 3. Divers 4. Infantile amnesia 5. State-dependent learning
Peak procedure was coined by ______. What is it?
Russell Church, Seth Roberts, and Warren Meck. A discrete-trial variation of a fixed interval schedule used to study timing in animals.
Scalar Expectancy Theory: What does the diagram look like?
See picture:
The internal clock
Seth Roberts: See 'Blackout' procedure. Warren Meck: The internal clock is like a stopwatch. It "runs" at a constant rate. - Methamphetamine speeds up the clock. - Haloperidol slows the clock. It can be stopped and restarted like a stopwatch.
Sarah Boysen, OSU
Showed that chimps can use digits to count fruits or other objects, and can even add two digits.
Kesner & DeSpain (1988)
Showed, using college students and rodents (separately) as subjects, that a combination of retrospective and prospective memory worked best.
Tolman, Ritchie, & Kalish (1946)
Sunburst maze and rats taking "shortcuts" Original training phase: Rats ran from A to G where they received food. A light was positioned behind the goal box so that it was pointed down the path from G to F. Animals learn very quickly to run straight to the food. Then in the next phase, a multitude of other pathways are brought in (the 'sunburst' look), and the original pathway was blocked. So did the animals choose the most similar pathway to the original? If they had, we would predict generalization, with the animal choosing the adjacent unblocked pathways. This is if the animal is learning a series of responses (S-R learning). But that's not what happened. It was cognitive learning. Results: The rats chose the most direct path to the food. However, this study has been controversial because in the original experiment there's a light present at the food location. So people criticize the study for this. HOWEVER, Ritchie (Tolman's graduate student) was able to definitively prove that the light was not an issue in a subsequent study.
Working Memory
Temporary "scratch pad" memory. Temporary retention of information that is needed for successful responding on the task at hand but not on subsequent (or previous) similar tasks. (Compare with reference memory.)
What is the Morris water maze?
The Morris water maze was a circular water-filled maze. The water was colored so that the rats couldn't tell it was water and to also conceal a hidden platform. The objective was to reach the hidden platform. Environmental cues were provided in the vicinity of the platform to get the rats attention. Normal rats would meander around the maze on their first try. By their 10th try they could move directly to the platform. In contrast, rats that had their hippocampus lesioned could not find the platform by the tenth try. Rats that had more training (days) had a lower latency to finding the submerged platform.
What happened in the chickadee study? ('saving room for dessert')
The chickadees that got no mealworms following some time after eating the readily available sunflower seeds, ate more sunflower seeds. The experimental group that received mealworms 30 minutes after eating a sunflower seed ate LESS sunflower seeds. They ate less because they preferred mealworms and were making room to eat those instead.
Clock Process
The clock process provides information about the duration of elapsed time. Components of the clock process: 1. Pacemaker 2. Switch 3. Accumulator A pacemaker that generates pulses at a certain rate (something like the beeps of a metronome). The pace- maker pulses are fed to a switch, which is opened at the start of the interval to be timed. Opening the switch allows the pacemaker pulses to go to an accumulator that counts the number of pulses that comes through. When the interval to be timed ends, the switch closes, thereby blocking any further accumulations of pacemaker pulses. Thus, the accumulator adds up the number of pulses that occurred during the timed interval. The greater the number of accumulated pulses, the longer the interval. The pacemaker/accumulator operates like an hourglass. When the hourglass is turned over, particles of sand pass to the bot- tom bowl, and the number of sand particles that accumulate there is a measure of elapsed time.
Scalar-expectancy theory of timing: Via the information-processing-theory of memory, what isn't part of the 'clock process'?
The comparator.
Decision Process
The contents of working and reference memory are then compared in the decision process, and this comparison provides the basis for the individual's response. For example, in the peak procedure, if the time information in working memory matches the in- formation in reference memory concerning availability of reinforcement, the decision is to respond. If information in working and reference memory does not match closely enough, the decision is to not respond. This mechanism produces a peak response rate close to the time when reinforcement is set up.
Memory consolidation theory
The establishment of a memory in relatively permanent form, or the transfer of information from short-term to long-term memory. Memory takes time to consolidate and can be disrupted during the consolidation period.
Memory Process
The number of accumulated pulses is then relayed to the memory process. The memory process thereby obtains information about the duration of the current stimulus. This information is stored in working memory. The memory process is also assumed to have information about the duration of similar stimuli in reference memory from past training.
Reconsolidation
The process of stabilizing or consolidating a reactivated memory. Presumably the disruption of this reconsolidation leads to modification or forgetting of the original memory.
What happens when the hippocampus is damaged? (Olton & Samuelson's 1976 experiment)
The rats cannot remember where they've been. Hence, they will go down arms they've already visited. However, they still know to go straight to the food reward. This is reference memory.
What is trace decay theory?
The theoretical idea that exposure to a stimulus produces changes in the nervous system that gradually and automatically decrease after the stimulus has been terminated.
Comparative Cognition
Theoretical constructs and models used to explain aspects of behavior that cannot be readily characterized in terms of simple S-R or reflex mechanisms. These mechanisms do not presume consciousness, awareness, or intentionality.
Chickadees "saving room for dessert"
Thinking of the future If you go to a pot-luck dinner (where each attendee brings something for the meal) and see many dessert options - you may decide to eat less of the main course so that you 'have room for dessert.' Animals also display this behavior. In an experiment with chickadees, the birds were given access to sunflower seeds for 5 minutes, followed after a delay by a dish containing mealworms. These birds much prefer mealworms to sunflower seeds. The experimenters wanted to find out if the chickadees would learn to hold back on eating sunflower seeds so as to have more room for mealworms. The delay between access to the sunflower seeds and the mealworms was 5 minutes during the first 15 trials and was then increased to 10 minutes and then 30 minutes. A control group was only given the sunflower seeds without the mealworm dessert later. Evidence of planning for the future was obtained at ALL delay intervals. The chickadees in the experimental group ate fewer sunflower seeds than birds in the control group in anticipation of the more delectable mealworms they were to get 30 minutes later.
Which experiment utilized the 'Sunburst Maze'? Describe the results.
Tolman, Ritchie, & Kalish's Sunburst Maze experiment. A maze which was shaped like a sunburst. The rat was first placed in an arena with one runway and allowed to familiarize with the surroundings. Then, multiple runways were placed into the arena and the originally familiarized runway was blocked off. The experimenters predicted generalization but... Ultimately, the rat ended up taking the shortcut to the food.
Brannon & Terrace (1998)
Trained rhesus monkeys to choose objects in ascending order of number.
What is episodic memory?
What? Where? and When? of memory.
Beatty and Shavalia (1980)
Working memory in the radial maze [Quick summary: Working memory lasts at least 8-12 hours.] To determine how long rats could remember where they have been, Beatty & Shavalia allowed rats to make four choices in the eight-arm radial maze in the usual manner. The subjects were then detained in their home cages for various periods up to 24 hours. After the delay interval, they were returned to the maze and allowed to make the remaining four choices (choices 5-8). As usual, an entry into an alley they had not previously visited was considered a correct choice. Delays of up to four hours after the first four choices did not disrupt performance. Longer periods of confinement in the home cage produced progressively more errors. These data show that spatial memory is not permanent. However, the memory can last for several hours. Beatty & Shavalia found that working memory lasted at least 8-12 hours.
The Scalar Expectancy Theory considers temporally controlled behavior to be the result of three independent processes: _____, _____, and _____,
[...]a clock process, a memory process, and a decision process.