Sniffy Software
Students may download a shareware version of the Sniffy the Virtual Rat software. Although this software is somewhat limited in its abilities, it is more than sufficient for the purposes of this assignment. Sniffy is a simulation, or model, of a real Sprague-Dawley strain laboratory rat in an operant chamber. Although the Sniffy software allows the user to experiment with classical and operant conditioning techniques, for this assignment students will not have to condition Sniffy; instead they will observe and record his behaviours for a set period of time.

Follow these instructions to download the Sniffy Demo software to a computer and make use of the software.

1. Go to http://www.wadsworth.com/psychology_d/special_features/ext/sniffy/index.htm, the Sniffy the Virtual Rat web site.

2. Select the "Download Sniffy" link on the left side of the web page.

3. Follow the instructions to download a Demo Version of Sniffy. Make sure to download the correct version for your computer's operating system (Microsoft Windows 95 or 98, Macintosh OSX, or Macintosh before OSX). For Windows users there is only one choice; click on the "Download" link to begin transferring the file to your computer. For Macintosh users you can select two download formats, MacBinary or BinHex. Depending on the configuration of your computer's software one or the other of the options should work. You may also need to download the program Stuffit Expander to use the downloaded Sniffy Demo; if you do not have the program Stuffit Expander, click the provided hypertext link to download this program.

4. Once you have downloaded the Sniffy Demo program to your computer, select the Sniffy Demo icon and follow the instructions for installing it on your computer's hard drive.

5. Open the Sniffy Pro Demo folder. Inside are several files. Open or select the "Sniffy Pro Demo" file to begin using the program. An operant chamber with a white rat (i.e., Sniffy) should appear on your monitor.

Observational Study Techniques
This assignment focuses on describing Sniffy's behaviour during a 1000-second (16.6 minutes) observation period. Technically, this is called a time-based sampling procedure. Despite the increasing use of technological recording devices in the field of psychology, experience in making behavioural observations is still essential whether you are working with patients in a hospital environment or animals in a highly controlled laboratory setting.

Although the observation procedure is relatively straightforward, analysis of time-based sampling procedures can be more complex. First of all, there is the issue of reliability of measurement. Are the observations consistent? This can be assessed by looking at whether two observers agree with respect to the behavioural observations. A simple way to measure agreement is to look at the percentage of observations in which two observers classify the behaviour in the same way. This is called an inter-observer reliability score.

Another method used to interpret observational data is the time budget. That is, what percentage of the time does Sniffy spend in different activities? When behavioural observations have been classified into mutually exclusive categories this measure is easily computed.

At a more complex level, one can ask how different behaviours relate to each other. For example, given that Sniffy is engaged in a bout of behaviour X, how likely is it that he will next do behaviour Y? One way to answer this question is to develop a transition matrix, in which behaviours are examined one at a time and then related to the behaviour in the next time interval.

More details about inter-observer reliability, time budgets, and the transition matrix will be discussed in class; ADDITIONAL HELP can be found on-line by taking this link.

Notes from the in-class discussion on what to include when WRITING THE ASSIGNMENT can be found here.

Collecting the Behavioural Data
Here's what you have to do to complete this part of the assignment. First, start a new Sniffy file (i.e., open the Sniffy Demo file or double click on the Sniffy Demo icon). Next, select "Preferences" from under the "File" menu. When the Preference window appears, set the"Animate Sniffy" option to the slowest position. This will reduce Sniffy's speed of behaviour to a manageable level for this assignment. Click "OK" to return to the Sniffy program.

Because you need to calculate an inter-observer reliability score you will need a partner for the data collection portion of this study. Anybody can serve as your partner for this assignment: a classmate, a friend, family member, etc. You and your partner are going to have to agree upon operational definitions of Sniffy's behaviour. I suggest that your partner spend some time watching Sniffy in his cage to see what sorts of behaviours he performs. Once you have identified the predominant behaviours, assign each behaviour a simple name and provide a brief definition of the behaviour. It is important that you and your partner agree upon the operational behavioural definitions so that your observations will be reliable when you begin recording Sniffy's behaviours.

For example, you may define the behaviour "Sniff" as, "twitching nose with all four feet on the floor", or "Walk" as, "rat moves across floor of cage using all four feet". I would recommend that you have somewhere between 5 to 9 behavioural categories. Fewer than 5 and you will not be able to categorize behaviours with any accuracy, whereas more than 9 may make it difficult record and analyze the data. Some behaviours, for example drinking from the waterspout, may be so rare as to hardly ever occur. One solution is to make your behavioural categories and then include an "other" category for the very rare or otherwise unclassifiable behaviours; you may have several rare or unusual behaviours listed within an "other" category.

Once you have your operational behaviour definitions you can begin recording data. You will observe Sniffy for 1000 seconds. During each 5-second interval, each of the observers is to independently record the dominant behaviour of that interval. When you are ready, start the timing and record on a sheet of paper with 200 spaces (a ruled sheet with different columns will suffice) the dominant behaviour exhibited during successive 5-second intervals. Make sure you stay synchronized with your partner (i.e., both you and your partner must watch Sniffy together and each of you will independently identify the behaviour you think Sniffy is doing most during each 5 second period).

After the 1000-seconds of observation are up you can exit the Sniffy program.

Working with your Partner
You need at least one partner to collect the behavioural data (i.e., the 200 observations). Students may work in groups of up to three people on this assignment. Groups of two work very well, although some students in the past have wanted two people for recording the data and one person to serve as a timer for the 5-second intervals. In any event, you will work on your assignment as a group. Therefore, your group will hand in one assignment for which everyone in the group will get the same grade. You should try to find your partner(s) from within the class.

Writing the Report
The assignment is to be written as if it were an APA-style journal article. That is, it should contain the following sections: abstract, introduction, methods, results, discussion, and references. Material on writing a paper in the APA format is available from the APA Style Manual which is in the reference section of all libraries on campus.

The APA website provides some useful tips for writing papers. Note that this information is not meant to be inclusive. Go to APA Style Tips for some advice on some common referencing and writing problems. You can also check the APA FAQ web page for answers to some commonly asked questions.

The following are just a few of my brief comments on the various sections of a scientific paper.

The abstract for this paper should be short and to the point. You probably do not need more than 100 words for a paper of this length.

The introduction of a paper provides background information for the reader, explaining the rationale behind the study, etc. I expect that this section will probably be quite short for your report, perhaps just a paragraph or two.

The methods section is extremely important as it allows other researchers to try to replicate your experimental design. You should include all details that are necessary for someone else to reproduce your study. Methods sections are often divided into subsections for the subject, apparatus, and experimental design. In this case, your subject is Sniffy. The apparatus is not only the computer you run the software on, but the simulated environment in which Sniffy "lives". In the experimental design you should indicate how all the data in the study is collected; this includes who collected the data, the operational behavioural definitions, the length of time you observed Sniffy, etc.

A results section should only contain results. That is, you provide the facts, but no additional speculation or interpretation of those results. Figures (a term taken to include tables, graphs, diagrams, and the like) may be used in the results section to help make your findings easily and succinctly understandable to the reader. Note that whenever you use figures you must number each one and include a brief caption explaining the content. For example, you might include a picture of Sniffy's operant chamber. You could label this Figure 1 and include a caption such as, "The simulated operant chamber in the Sniffy Demo software." Note that tables, graphs, and figures can not substitute for explanatory text in the results section (i.e., a couple of graphs alone will note be enough). If you use figures make sure that you refer to them in the section's text and that you put them in context. For example, you might write something like the following: "My results show that Sniffy demonstrated a marked preference for engaging in a period of walking after sniffing, as seen in the transition matrix in Figure 3." It is very important to direct the reader to the appropriate figure at the right time while they are reading your paper.

If you use figures you may either use a computer program to produce them (if you are technically savy) or you may make your figures by hand, provided that you are neat and tidy (e.g., use a ruler to draw straight lines instead of "free-handing" it). Figures may either be put in the body of the text of your paper or, if you prefer, it is perfectly acceptable to put all the figures in an Appendix section at the end of the paper. Again, make sure that you carefully label each figure so that the reader knows its relevance.

In the discussion section you interpret the results and comment on what they might signify. What can you conclude about Sniffy from the data you have collected? What is interesting about the data? Explain your results in the discussion section.

Finally, you should must a reference section. Here you list, in proper APA reference style, any and all sources that you cited in your paper.

For more information on writing the different sections of a scientific paper please take this link to the excellent website on Reading the Research. While this site was constructed for introductory students it is very relevant to this assignment.

Required Elements
There are several components that I do require for this assignment:

1. You must use APA format when writing your paper.
2. You must include your operational behavioural definitions.
3. You must include your inter-observer reliability score, the time budget, and the transition matrix.
4. You need to find at least one journal article that is somehow relevant or related to your study and cite it appropriately in your paper. This has to be an actual journal article; an abstract from an electronic library database, for example, doesn't count. You will have to actually read the articles to make sure that it is relevant to your study. The article you cite might, for example, relate to the observational techniques you used (e.g., the use of observational studies with rats, pigeons, people, etc), the methodology (e.g., the use of computer simulations in psychological studies, etc.), the interpretation of results (e.g., the use of inter-observer reliability scores, time budgets, transition matrixes, etc.), your interpretation of the results (e.g., comparing your results with results from live rats, etc.), or the like. These are only a few ideas; there are many ways that you can work citations into your paper. If you have specific questions, please ask me.
5. I am requiring you to hand in a photocopy or printout of the first page of each article that you reference; please staple them to the back of your paper.
6. The paper must be typed in 12 point font, double spaced, with 1 inch margins. Include a title page with your name(s) and student ID(s); only put your name(s) and ID number(s) on the title page.
7. Staple the "raw data" (the sheet on which you recorded the 200 behavioural observations) to the back of your paper.

Submitting your Assignment
Your paper must be turned in to the instructor during the 5 June 2003 class. A 10 percent penalty will be applied per day that your assignment is late, counting weekends, starting immediately after class on 5 June. Do not try to turn assignments in by sticking them under the instructor's door! If you do not turn your assignment in to the instructor in class on 5 June, take it to the Department of Psychology General Office (BSP-217) and ask one of the office staff to put it in my mailbox.

A Final Word
This is not meant to be a gigantic project. Keep the sections of your paper tight and focussed on the essential elements. Use the same technique of editing drafts of your work that I recommended for the second assignment.

URL: www.psych.ualberta.ca/~msnyder/p281/assig/sniffy.html
Page created: 30 July 2002 --- Last modified: 19 May 2003