Excerpted from "Sniffy, the Virtual Rat: Integrating Sniffy into a Learning and Behavior Course" by Michel R. Snyder. In Introduction to Learning & Behavior: Instructor's Manual with Test Bank by Russell A. Powell. For marking key, see instructor's manual.

Part 1 (Sniffy manual: Chapter 3)

Question 1:

Follow the steps in Exercise 1 (pp. 21-23) to magazine train Sniffy. Save your magazine-trained version of Sniffy. Next, following the steps in Exercise 2 (pp. 23-26) shape Sniffy to lever press. Make sure you save a new copy of Sniffy once he is fully conditioned to press the lever.

Answer the following question: Throughout the lever pressing shaping process, why should you try to deliver a pellet of food as soon as possible after Sniffy does the desired behaviour?

Question 2:

It is also possible to use autoshaping to condition Sniffy to press the lever. Load the first saved version of Sniffy from Exercise 1 (i.e., Sniffy is magazine trained, but not trained to lever press). From under the "Experiment" menu select "Isolate Sniffy." This speeds up the rate at which program time passes. Pay attention to the Cumulative Record and the Operant Association window (specifically the Bar-Sound measure). After a while the Cumulative Record will indicate that Sniffy has pressed the lever and received a food pellet. Eventually, Sniffy will press the lever more frequently and you will see the Bar-Sound graph rise, indicating that Sniffy is learning the appropriate association.

Compare the Cumulative Record from when you shaped Sniffy by hand (i.e., in Exercise 2) with the Cumulative Record of the autoshaped Sniffy. How do they differ with respect to the development of the Bar-Sound association and in terms of the amount of program time required to produce a fully conditioned rat?

Question 3:

Sniffy is a simulation, or model, of a live rat. As such, he lacks certain cognitive features of real rats. Given this, why might it be much harder for autoshaping of the lever pressing response to occur in a real rat, especially given the longer time frame over which autoshaping takes place?

Question 4:

Complete Exercise 5 (pp. 34-38) to extinguish Sniffy’s lever pressing response and answer the following: How many responses did Sniffy make after the extinction criteria was imposed but before he actually extinguished his lever pressing response and how long (in program time) did this take?

Attach to your assignment a printout of the portion of Sniffy’s Cumulative Record showing the five minute period (of program time) before and after the extinction burst. On this Cumulative Record circle (or otherwise identify) the extinction burst.

Question 5:

Complete Exercise 6 (pp. 38-40) to extinguish Sniffy’s lever pressing using the alternate technique in which the "Mute Pellet Dispenser" option is turned off.

Answer the following: How many responses did Sniffy make under this condition before he extinguished his lever pressing response and how long did this take? How does this differ from your results in Question 4? Why?

Question 6:

Does the difference in extinction time and number of responses made under the two extinction conditions (i.e., pellet dispenser sound muted or not muted) seem realistic given that Sniffy does not receive any food once the extinction protocol is put in place in either case? What does this imply about the strength of secondary reinforcers in controlling behaviours?

Complete Exercise 7 (pp. 40-42). Make sure you use your saved version of Sniffy from Exercise 5. Also, save a version of Sniffy after his responding has extinguished following the spontaneous recovery procedure. You will need this for Question 8 below.

How many responses did Sniffy make and how much program time was required for him to reach the extinction criterion?

Question 8:

Using the version of Sniffy you saved in Question 7, give Sniffy another time out so that he will show spontaneous recovery and extinction a second time. Finally, give Sniffy a third time out and allow him to extinguish his responding again. Print the Cumulative Record showing the initial extinction and the two spontaneous recoveries following the two time outs that you delivered. (Follow instructions described in step 16a on p. 36 to print the Cumulative Record.) Your Cumulative Record should show the initial extinction of Sniffy’s bar pressing, followed by Sniffy’s response patterns after the three time outs. Attach your Cumulative Record to your assignment and use the information on it to complete the following question:

Calculate the number of responses made and the time required for Sniffy to extinguish lever pressing for the initial extinction phase and for each of the three subsequent extinguished spontaneous recoveries. Using this data, fill in the line graphs below. Be sure to add the numeric scale to the Y-axis of each graph.

Key: EXT = initial extinction trial, SR1 = first spontaneous recovery, SR2 = second spontaneous recovery, and SR3 = third spontaneous recovery.

Following the instructions in the Sniffy Pro text, complete Exercises 8, 9, 10, 11, and 12 with the aim of achieving the following:

1. Condition Sniffy to VR-5, VI-5, FR-5, and FI-5 reinforcement schedules.
   
   
2. Condition Sniffy to VR-25, VI-25, FR-25, and FI-25 reinforcement schedules.
   
   
3. Condition Sniffy to VR-50, VI-50, FR-50, and FI-50 reinforcement schedules.
   
   

Then, complete the following:

Calculate the response rates for each of the four schedules at each of the four rates of reinforcement. To calculate the response rate you need to find the slope of the Cumulative Record for each reinforcement schedule. This will require you to print the Cumulative Record for each schedule, or to use the cut and paste functions (see pp. 42-43 for instructions) to transfer representative samples from the Cumulative Records into a word processor for printing. The slope of a line is its "rise" divided by its "run", that is, the increase in the vertical distance divided by the horizontal distance covered. For example, if Sniffy presses the lever 75 times in 10 minutes then his response rate is 75 resp./10 min., or 7.5 responses/minute.

Complete the calculations in the chart below. Also, attach a copy of the Cumulative Records that you used to calculate the response rates to your assignment. On the Cumulative Records indicate what sections you used in your calculations.

Next, plot the information on the line graph below. The X-axis depicts the value of the reinforcement schedule (i.e., 5, 25, and 50) and the Y-axis the rate of response. Make sure you fill in the numeric scale on the Y-axis. Be sure to label the four lines with the appropriate schedule name (i.e., VR, VI, FR, and FI).

Briefly explain the differences in Sniffy’s response rates for the four schedules.

Question 10:

Complete exercise 13 (pp. 58-59) in the Sniffy text. Place Sniffy on extinction for VR-25, VI-25, FR-25, and FI-25 schedules and answer the following question:

Determine how many responses and how much time it takes Sniffy to extinguish on each of the four schedules. Rank the schedules from easiest to hardest to extinguish, and explain your results.

In preparation for the questions in this section read the background material in chapter 5 (pp. 61-73) in the Sniffy Pro text, and then complete Exercises 14-19 in Sniffy Pro.

Question 11:

Answer the following: for the generalization gradient tests in Exercises 15, 17, and 19, why do you need to restart from a saved version of Sniffy for each of the nine tone frequencies that you present for the complete generalization gradient test?

Question 12:

Print out and include the DS Response Strength graphs from exercises 14, 16, and 18 (label the graphs clearly). Explain the differences between these three graphs.

Question 13:

Produce a graph of the three generalization gradients from exercises 15, 17, and 19 and include it with your assignment.

Question14:

Answer the following: What would the stimulus generalization gradient look like if you trained Sniffy on an S+ of 2.0 kHz and two S- of 1.75 and 2.25? (Unfortunately, you can’t actually do this with the Sniffy software.) Explain this in terms of Spence’s theory of peak shift.

Question 15:

Complete the following:

1. Run discrimination learning trials for the following pairs of S+/S- stimuli: 1. S+ = 2.0 kHz, S- = 1.0 kHz; 2. S+ = 2.0 kHz, S- = 1.5 kHz; 3. S+ = 2.0 kHz, S- = 1.75 kHz. (You can follow the general instructions from Exercise 18 to complete these S+/S- learning trials; simply replace the values in step 4 with those listed here.)
   
   
2. Run stimulus generalization tests for the nine tones (i.e., 1.0 kHz to 3.0 kHz) for each of these three discrimination learning sets. (You can follow the general instructions from Exercise 19 to complete these stimulus generalization gradients.)
   
   
3. Generate a graph, of the sort depicted on p. 81 of Sniffy Pro, showing the generalization gradients for these three discrimination learning sets; also include the generalization gradient from Exercise 15 (i.e., S+ = 2.0 kHz, no S-).
   
   
4. Explain this graph in terms of behavioural contrast and peak shift. Use data from the DS Response Strength graphs to help explain the peak shifts seen in the data.
   

Question 16:

Complete Exercises 20, 21, and 22 (pages 104-112) on acquisition, extinction, and spontaneous recovery, respectively. Follow the instructions in the Sniffy text with one exception. For these exercises, type "20" in the text box located to the right of the Present Each Trial Type instead of "10"; that is, you will present the CS-US pairing 20 times to Sniffy. Use the copy of the VR25 file located on the Sniffy Pro CD for questions in this section. Follow the instructions on p. 105 to use this VR25 file.

Print both the Suppression Ratio and CS Response Strength graphs after you have completed exercise 22 and attach them to your assignment. Make sure you label the graphs so that they can be identified by the marker.

Question 17:

As it turns out, it does not matter whether you use the tone or the light as the CS for these studies; both stimuli give the same CS Response Strength numbers. In other words, the tone and the light are treated by Sniffy as being equally salient and relevant. Would you expect such an occurrence from real rats? Why? Explain your answer.

Question 18:

Complete Exercise 23 (pages 112-119). As with the exercises in Question 16, for both the low and high CS intensity conditions present 20, instead of 10, CS-US pairings. Follow the instructions on pages 115-119 to produce a line graph of the CS response strength for the three CS conditions (i.e., low, medium, and high intensity tone).

Attach the CS Response Strength graphs for the three intensities of CS to your assignment. Ensure that you title and label the graphs appropriately.

Question 19:

Exercise 24 (pp. 120-122) deals with the influence of US strength on the learning of the CS-US association in classical conditioning.

As in Questions 16 and 18, run 20 instead of 10 CS-US pairings.

Using the instructions from Exercise 23, produce a line graph of the CS Response Strength of the three levels of US intensity (i.e., low, medium, and high) with a medium intensity CS. Label the graph completely. Attach the graph to your assignment. Using this graph, answer the following: For the low strength US, is there a point of diminishing conditioning? Explain this in terms of habituation. According to your results, what is the point of diminishing conditioning for Sniffy?

Question 20:

How is habituation of a low-intensity US like or unlike extinction of a medium-strength US? What happens in terms of the Suppression Ratio and the pain sensitivity measure in both of these conditions?

Question 21:

Suppose you only have a record of the Suppression Ratio of Sniffy’s performance and do not know what conditioning situation Sniffy has been exposed to. Using only the behavioural data of the Suppression Ratio, is it possible to determine if Sniffy has been learning an association between a low-intensity US and a medium-intensity CS, or if he had a medium-intensity CS extinguished? Explain your answer.

Complete Exercise 25 (pp. 124-130). Because of the random variability built into Sniffy, you can sometimes get some rather peculiar results in the final testing phase. For this reason, students should test the appropriate stimuli three times, rather than once. The easiest way to do this is to set the Present Each Trial Type to "3", instead of "1". This will allow you to test the CS of interest three times and take an average value.

Print the Suppression Ratio and CS Response Strength graphs from the Compound-Conditioning Experiment Condition (pp. 124-127) and the Compound-Conditioning, Separate-Pairing Control Condition (pp. 127-130) and attach them to your assignment. Label the graphs appropriately so that the marker can identify them.

In the space below, provide the values of the averaged Suppression Ratios (i.e., the average of the three test trials you conducted for each stimulus) for the tested stimuli from the two parts of Exercise 25.

Compound Conditioning

CS_ML: ___________ CS_MT: ___________

Separate Pairing

CS_ML: ___________ CS_MT: ___________

Question 23:

Answer the following: In experiments with real animals researchers might use a technique similar to the one used in Question 22 (i.e., presenting test trials more than once) to control for an animal’s response variability. The drawback of this approach is that the animal may begin to extinguish to the CS because it is not followed by the US on testing trials. What is an alternative technique that could be used to control for response variability?

Question 24:

Complete Exercise 26 (pp. 130-135) on blocking. As in Question 22, test each CS three times. In the space below, fill in the appropriate values based upon the averaged suppression ratios from your data. Note: essentially, you are being asked to provide the actual values from your data for the Stage 3 section of the chart on the top of page 131. Print, label, and attach to your assignment the appropriate Suppression Ratio graphs corresponding to the numerical values that you provide below.

Blocking Experimental Condition

CS_ML: ___________ CS_MT: ___________

Compound Conditioning Control

CS_ML: ___________ CS_MT: ___________

Question 25:

Complete Exercise 27 (pp. 135-138). Print and label the Suppression Ratio and CS Response Strength graphs and attach them to your assignment. How might your results have differed if you had used a low-intensity light and a medium-intensity tone, instead of the high-intensity tone that you did use? Support and explain your answer.

Question 26:

Complete Exercise 28 (pp. 138-141) on the overexpectation effect. Print and label the Suppression Ratio and CS Response Strength graphs and attach them to your assignment. What would happen if you had used a high-intensity shock instead of a medium-intensity shock in stage 3 of the chart on page 139? Provide support for your answer. Hint: the easiest way to answer this is to actually test the condition using Sniffy.

Question 27:

Complete Exercise 29 (pp. 144-148). Print and label the Suppression Ratio and CS Response Strength graphs from Exercise 29 and attach them to your assignment

Question 28:

Complete Exercise 30 (pp. 148-152). Print and label the Suppression Ratio and CS Response Strength graphs and attach them to your assignment. Consider the results from Exercises 29 and 30. Both provide measures of conditioned inhibition. Discuss the importance of both of these techniques in explaining the working of conditioned inhibition.

Question 29:

Complete Exercise 31 (pp. 154-160). Then run the following exercise.

Open the VR25 file, then use the Save As command to give the file a new name (e.g., Ex31-SPCMOD). Choose the Change Nature of the Association command from the Experiment menu and select the S-S association for CS used as US; click OK. Choose Design Classical Conditioning Experiment from the Experimental menu. Under Stage selection, set the Interval Between Trials to 5 minutes and the Preset Each Trial Type to 5 times. In the First Stimulus section of the dialogue box, select both medium-intensity light and medium-intensity tone. In the Second Stimulus section of the dialogue box, select none. Click the New Stage command button (you should now be editing Stage 2, Trial Type A). Set the Interval Between Trials to 5 minutes and the Preset Each Trial Type to 10. In the First Stimulus section of the dialogue box, select the medium-intensity tone and in the Second Stimulus section of the dialogue box, select the medium-intensity shock US. In the Stage section of the dialogue box, click the New Stage command button (you should now be editing Stage 3, Trial Type A). Set the Interval Between Trials to 5 minutes and the Preset Each Trial Type to 3 times. In the First Stimulus section of the dialogue box, select the medium intensity light and in the Second Stimulus section of the dialogue box, select None. Click the Save command button and then execute the Run Classical Conditioning Experiment command.

Print the CS Response Strength and Suppression Ratio windows and attach them to your assignment. Label the graphs clearly so the marker can identify them.

How do your results compare to those in the Unpaired Control Condition and Experimental Condition of Exercise 31 in the Sniffy text? In Stage 1 of the Experimental condition the light is designated as the CS and the tone as a US. That is, the light turns on by itself to be followed at the end of the presentation period by a simultaneous presentation of the tone. Consider the difference in the modified exercise you have just run. In this case the tone and light come on simultaneously for the entire CS presentation period. However, your results are probably quite different from the Experimental condition results. Comment on your results in terms of sensory preconditioning. What associations are being formed? Do the results of this modified exercise agree with the sorts of results you would expect from live animals exposed to various forms of sensory preconditioning?

Question 30:

Complete Exercise 32 (pp. 160-163). Exercise 32 demonstrates higher-order conditioning, specifically what is called second-order conditioning. Third-order conditioning can also be produced. In third-order conditioning an additional CS is subsequently associated through pairing with the second-order CS. The following instructions will allow you to conduct a third-order conditioning test with Sniffy. Fourth and even higher-order conditioning is possible with live animals, but in practice it becomes progressively more difficult to form the higher-order associations. With Sniffy we are limited by the number of stimuli available in the operant chamber (i.e., light, tone, bell) to testing third-order conditioning.

Follow the instructions for Stage 1 of Exercise 32. Then make the following changes. Click the New Stage button (you should be in Stage 2, Trial Type A) and set the Interval Between Trials to 5 minutes and Present Each Trial Type to 5 times. In the First Stimulus section of the dialogue box, select the Bell and in the Second Stimulus section of the dialogue box, select the medium-intensity Tone as a CS Used as US. Click the New Stage button (you should be in Stage 3, Trial Type A) and set the Interval Between Trials to 5 minutes and the Present Each Trial Type to 5 times. In the First Stimulus section of the dialogue box, select the medium-intensity Light and in the Second Stimulus section of the dialogue box, select the Bell as a CS Used as US. Click the New Stage button (you should be in Stage 4, Trial Type A) and set the Interval Between Trials to 5 minutes and the Present Each Trial Type to 3 times. In the First Stimulus section of the dialogue box, select the medium-intensity Light and in the Second Stimulus section of the dialogue box, select None. Click the Save button and then select Run Classical Conditioning Experiment from the Experiment menu.

Print and include your CS Response Strength and Suppression Ratio graphs in your assignment. Label them clearly so the marker can identify them.

Answer the following: Do your results agree with what you would expect for third order conditioning? Why or why not?

Question 31:

Read pp. 165-168 and then complete Exercises 33-34 (pp. 168-174). Explain your results in terms of S-S and S-R theory. More specifically, how does the nature of the association when the CS is paired with the US determine the experimental results?

Question 32:

Read the material on pp. 187-188 and run Exercise 38 (pp. 189-194). Answer the following:

Why does a suppression ratio not provide useful data for the interpretation of background conditioning (i.e., conditioning of contextual stimuli)?

Print, or examine on the computer screen, the Cumulative Record of conditioning trials 1-5, 21-25, and 46-50 for each of the three US strength conditions (i.e., low, medium, and high intensity). Calculate the average time that the US suppresses Sniffy’s bar pressing for the three periods of each of the intensity conditions. Remember that the Cumulative Record represents ten minutes of time by the space between two solid vertical lines. To calculate the time of suppression, measure the length of the Cumulative Record’s line that remains flat following the termination of the US (note that each pen movement upwards on the Cumulative Record represents one bar press by Sniffy). Then measure the length of Cumulative Record between the ten-minute marker lines. Divide the length of your first measure by the length of the second and multiply by 10 to produce a value in minutes. Plot your averaged results on the graph below. Fully label your graph.

Question 33:

Read the material in the Sniffy text related to exercise 39 (pp. 194-197) and complete this exercise.

Explain why the CS pre-exposure effect slows the acquisition of learning. How is the CS pre-exposure effect similar to habituation?

Question 34:

Read the material in the Sniffy text related to exercise 40 (pp. 197-200) and complete this exercise.

Using the graph shown on the bottom of page 199, propose an explanation for why both the CS pre-exposure and US pre-exposure effects result in an asymptotic maximum CS Response Strength that is less than that for the control condition. Also, see if you can theorize a reason for why the US pre-exposure effect has less of an effect on the speed of conditioning than the CS pre-exposure effect. Note that the data presented in the graph is generated from the Sniffy software and, as the text states on p. 200, the data may not be representative of live rats. (If you are particularly keen you may wish to construct your own graph; note that you should probably run Sniffy at least three times through each of the conditions [i.e., control, CS pre-exposure, and US pre-exposure] and average the results for each trial to control for variability.)