1
B130
CRICKETS: WHAT, WHEN, AND WHY’D THEY DO THAT? OBSERVING AND DESIGNING STUDIES OF COMPLEX BEHAVIOR
Observing animal behavior seems simple: just record what happens. In practice, it is rarely that easy. If you only wanted to record your impressions, you could write down just about anything. If, on the other hand, you want to understand how or why an animal behaves the way it does, something that other observers would also find if they repeated your experiment, you have to be more precise.
Goals:
1) Determine whether there are individual differences in the behavior of crickets, or whether they are all genetically programmed to perform the same behaviors.
2) Learn how to observe, define, categorize, and collect complex behavioral data.
3) Learn important statistical techniques used in data analysis.
The laboratory has two parts. In Part I you will observe the behavior of captive crickets and describe qualitatively what you observe. In Part II you will quantify your observations to test specific hypotheses, using chi-square contingency tables and goodness of fit tests.
CRICKETS PRE-LAB ASSIGNMENT:
There are FIVE items that comprise your pre-lab assignment:
1) Remember to bring your calculator to lab! Put it somewhere so that you won’t forget it!
2) Read over Part I of your lab to learn about crickets and cricket behavior. During the first portion of this lab you and your partner will first watch crickets to create a detailed “menu” of behaviors on each page. For each behavior you will assign a “code” (abbreviation) of letters. You will need to familiarize yourself with the different kinds of behavior crickets can exhibit and how to categorize these behaviors.
3) Carefully read over the instructions for your qualitative and quantitative observations and look over the flow chart (pg. 11) so that your experiments will run smoothly once you get into lab. You and your lab partner will be recording data for Experiments A and B and analyzing class data for Experiment C. It is crucial that you understand how to run your experiments so that you have good data to analyze.
4) For each experiment on your data sheet please indicate 1) what your experiment is testing, 2) your null hypothesis, and 3) what you can conclude if you reject your null hypothesis. Do this before you come to class!
You must understand what your outcomes might signify. That is, you must understand why you’re conducting these experiments and how to interpret your results. What aspects of cricket behavior are you testing? Scientists must be skeptical. You must initially assume that no pattern exists (that crickets are just acting randomly) and modify your conclusions only when you have evidence that a pattern does indeed exist. A null hypothesis is what you would expect to see if no pattern exists. If you can reject your null hypothesis, than your experiment provides evidence for a particular pattern of cricket behavior.
2
5) Pre-lab practice analysis of data:
How will you know whether your data shows a real pattern in cricket behavior or whether the differences you see are just random differences? Scientists use statistics to determine whether patterns are significant or not. During lab you will conduct two types of chi-square (2) tests: chi-square contingency table and chi-square goodness of fit. You will use these tests to determine the probability (P-value) that your results are random (due to chance.) It is important to understand the basics of these tests, so that you can work through them quickly during lab.
Please read through examples 1 and 2 in the statistical analysis appendix in the back of your lab packet, then run statistical tests using the sample results in your prelab worksheet. As you work through the analysis, answer the questions on your pre-lab worksheet. You will be turning these in at the beginning of the lab period.
revised Apr/2013 JMN
3
BIOL 130 NAME __________________________
CRICKET BEHAVIOR PRELAB PRACTICE DATA ANALYSIS
Read over the statistical appendix at the end of your lab packet, paying particular attention to examples one and two. Use this information to help you with the following sample problems.
2 CONTINGENCY TABLE TEST
1. Experiment A in your cricket lab tests the following hypothesis: The behavior of Cricket 1 will be significantly different from the behavior of Cricket 2 when they are interacting with each other. What is your null hypothesis?
2. The following table provides a sample data set, similar to what you might get after 15 minutes of observation.
3. Calculate the row and column totals for your sample observed data table above and construct a second data table of expected values.
4. Look closely at your expected values for “neutral” behaviors. Are they < 5? If so, then you must ignore the “Neutral” behavior category, because you have too few data to test this category properly. Eliminate the neutral category in both your observed and expected tables and re-calculate your row and column totals. Your data tables will now only contain 4 cells.
SAMPLE RESULTS FROM CRICKET EXPERIMENT A:
Observed Data:
Solitary
Aggressive
Neutral
Cricket 1
20
2
2
Cricket 2
15
16
1
Expected Data
Solitary
Aggressive
Neutral
Cricket 1
Cricket 2
Expected Data
Solitary
Aggressive
Cricket 1
Cricket 2
4
5. Calculate the chi-square value (please show your work)
Chi-square = sum of all (obs-exp)2
Exp
6. Calculate the degrees of freedom (df):
(number of rows – 1) x (number of columns – 1)
7. Use the chi-square table in the appendix to determine the probability level (P-value) of your chi-square. What is the probability that your results could be due to random chance? (check one)
_____greater than 0.05? (your chi-square value is less than the number in the P = 0.05 column)
or
______less than 0.05? (your chi-square value is more than the number in the P=0.05 column)
8. Do you support or reject your null hypothesis for the sample experiment?
9. What does this tell you about your crickets’ behavior in the sample experiment?
Continue to the next page for practice with Goodness of fit tests
5
Goodness of Fit 2 test
10. Experiment C in your cricket lab uses the class data to determine whether crickets overall have predictable behavior. Did most of the crickets that you examined in experiments A and B act the same way in both trials? Did most change their behavior from one trial to the next?
The null hypothesis in this experiment is that cricket behavior is unpredictable. Some may act the same and some may act different. We cannot predict which.
Your class data might look something like the table below:
OBSERVED DATA
SAME (number of groups that accept Null Hypothesis in Exp. B)
DIFFERENT (number of groups that reject Null Hypothesis in Exp. B)
GRAND TOTAL
8
12
20
11. If there is no significant difference between the number of crickets that acted the same and the number of crickets that acted differently, you can calculate your expected data simply by dividing the total number of crickets tested (20) by the number of categories (2). This gives us the number of crickets in each category for the expected values for the null hypothesis (crickets are equally likely to act the same or act differently). Calculate your expected values and insert them in the table below.
EXPECTED DATA
SAME (number of groups you would expect to accept Null Hypothesis in Exp. B)
DIFFERENT (number of groups you would expect to reject Null Hypothesis in Exp. B)
GRAND TOTAL
12. Calculate the chi-square value (please show your work)
Chi-square = sum of all (obs-exp)2
Exp
13. Calculate the degrees of freedom (df):
= (number of categories) – 1.
6
14. Use the chi-square table in the appendix to determine the probability level (P-value) of your chi-square. What is the probability that your results could be due to random chance? (check one)
_____greater than 0.05? (your chi-square value is less than the number in the P = 0.05 column)
or
______less than 0.05? (your chi-square value is more than the number in the P=0.05 column)
15. Do you support or reject your null hypothesis for the sample experiment?
16. What does this tell you about the predictability of cricket behavior overall in the sample experiment?
7
Part I: Qualitative Observations
Background:
How and why to male crickets “chirp”?
You will observe Acheta domestica, the house cricket. Males compete for dominance by ritualized combat and singing (Figure 1). Dominant males overpower opponents in wrestling matches. Following a fight, dominant crickets return to singing while subdominant males no longer sing. Females are attracted to singing (and thus dominant) males as sexual partners.
Figure 1. House cricket (a) male “singing”, (b) the sound-producing mechanism is located on the forewing of the male. Forewings have thickened structures called tegmina (s. tegmen). (c) detail of basal portion of tegmina, showing how the plectrum of one wing fits into the files of the other to produce vibration of the wing membranes.
Categories of behavior:
Behavior may occur in a solitary or social context (Table 1). Social behavior may or may not involve aggression. The three major categories you will observe are: solitary, aggressive, and neutral behaviors (Table 1). Within each category, you are likely to see many different types of specific behaviors. We have listed some for you in Table 1, but you may find others!
Table 1: An overview of some common behaviors of the male house cricket above, Acheta domestica.
SOLITARY
AGGRESSIVE
Aggressive and submissive social behavior
NEUTRAL
Non-aggressive social behavior
Locomotion
Chirping
Crawling over or under another
Self-grooming
Side-by-side display
Tapping another with antennae
Digging
Grappling
Resting next to or on another
Resting
Flipping
——-
Exploratory
Head to head contact
——-
IMPORTANT: you will need to make some arbitrary decisions with your partner about how you classify particular behaviors into one of the three categories. For example: is exploratory behavior considered social if a cricket happens to bump into another? Is singing alone social? Answers: probably “no” to the first question and “yes” to the second.
8
Instructions:
Part I: Qualitative observations of behavior (work in groups of 2)
1. On three separate pages, create a datasheet for three different categories of behavior: Solitary, Aggressive (Agonistic), and Neutral (as indicated below).
2. QUIETLY enter the lab. WHISPER, NO VIBRATION (do not slap notebook on table or bump table).
3. Get out your prepared behavior data sheets and turn to Table 1, overview of cricket behavior on page 7.
During the first ten minutes:
Spend 8 minutes only watching the crickets in plastic cage in the center of your table.
Spend the remaining two minutes quietly visiting another group to observe behaviors that you have not yet seen.
As you watch, record all qualitative descriptions of behavior:
1) Describe and sketch behaviors on the appropriate page(s) and make a “code” for each (e.g. W = walking). You and your lab partner will need to agree on which behaviors you consider aggressive, solitary or neutral. Share your codes with each other as well.
2) Note the variation of any behaviors that you see more than once. How does it vary within a single individual? E.g what different kinds of grooming behavior do you see?
3) Watch closely! Behavior can be subtle; animals rarely do “nothing” Watch antennae, mouthparts, etc.
4) Spend the final 2 minutes of the session recording behaviors of crickets on another table (aim to find aggressive behaviors, such as chirping, aggression, or other behaviors you haven’t seen).
You should now have a set of brief behavioral descriptions and a list of codes for each behavior. It is very important that you and your partner(s) share your information, such that you agree on the behaviors that you will watch for, when you proceed to Part II.
Record each behavior “code” and what it stands for (i.e. W = walking) at the top of your data tables for Experiment A and Experiment B (your 60, 15-second time interval tables). You will refer to these “codes” as you collect data in Part II.
When you have completed the above information, have an instructor check your lab before proceeding to Part II. Now that you have described the major behaviors of crickets, you are ready to record a sequence of behaviors over time.
9
Part II: Quantitative observations and time interval recording
You will run two 15-minutes experiments. Each involves a pair of crickets. Examine the flow chart carefully, and read all instructions before you begin. The crickets have been housed in two separate cages. Animals from different cages will have a different color painted on their back. Instructions for Cricket Experiments A and B:
Quantitative observations. You will record quantitative observations for 60, 15-sec time intervals on the data sheets at the end of this lab. One of the many advantages of the data tables that you will complete is that they will preserve the timing and sequences of behaviors for later analysis.
1) At the top of each table write down the names of the observers. Please read and understand what the table entries mean, in the description below.
SAMPLE DATA TABLE
Behaviors Observed
EXPERIMENT A
Cricket 1 (color #1)
Cricket 2 (color #2)
15-sec period
Solitary
Aggressive
Neutral
Solitary
Aggressive
Neutral
1
R
Ch
2
R
Lo
Ant
3
R
Lo
Ch
(continued to 60!)
R
Dig
Ch
60
R
Lo
Ch
When you observe your crickets, you will put a “code” for the type of behavior that was observed during each 15-sec time period. In the example above, Cricket 1 rested (R) almost the entire time, while Cricket 2 engaged in a variety of behaviors including locomotion (Lo), digging (Dig), chirping (Ch), and antennae tapping (Ant). Often, Cricket 2 did more than one type of behavior in the 15-second period, as in the last three entries in the table. However, please note that within each category (Solitary, Agressive, Neutral) only one behavior was scored. Each data point represents the first behavior observed for that category (i.e. “Ch” was the first agonistic behavior you saw during the 3rd 15-second interval for Cricket 2).
You may wonder why each category should be scored only once within a 15-sec time interval. What if a cricket exhibits two aggressive behaviors (i.e. chirps and antenna taps within the same 15-sec interval)? Why only write down the first chirp and ignore the second aggressive behavior (the antenna tap)? Remember: you are not a human video camera! Believe it or not, you are not really “throwing away data,” if you miss some behaviors. Behavioral studies, like any scientific investigation, provide a way of systematically sampling what is going on. If chirping, for example, really is a more common behavior for Cricket 2 (and not for Cricket 1) then you will find that chirping occurred in multiple 15-second intervals for Cricket 2 (and you do! See sample data table). Sampling at 15-sec intervals also gives you time to glance back at your notebook to record behaviors, before starting your next 15-sec interval.
Your statistical test will compare the relative differences in chirping between the two crickets across the 15-sec intervals. If the crickets are systematically sampled the same way, you are equally likely to miss a few behaviors (i.e. chirps) for each individual: yet you will still be able to detect a relative difference between the two (Cricket 2 chirps more than Cricket 1).
10
You will run two 15-minutes experiments. You and your partner will each watch one cricket as they meet for the first time. Using the data sheets provided, you will record behaviors of the cricket you watch in each of the 60, 15-second intervals.
IMPORTANT: You will not record everything! Within each of 15-sec interval (listen for the computerized beep to mark the start of each interval), record the behavior “code” for the FIRST behavior that your cricket exhibits in EACH of the 3 categories. That is, you will have a maximum of 3 data points per 15-sec interval (no more than 1 solitary, 1 aggressive, and 1 neutral behavior); most intervals will have less than 3 behaviors. You will be recording data for only your cricket.
Conduct Experiment A by QUIETLY (without bumping) joining the single-male cages together and removing the lids. Give your crickets a moment to adjust. Synchronize with your partner to start collecting data. Relax. Crickets usually don’t do much initially! Use the first convenient “beep” to mark the beginning of your first 15-sec time interval and record from there.
After completing the 60 intervals (15 minutes) very gently separate your crickets into their cages. One of you take your cricket and exchange places with a nearby group to begin Experiment B (see flow chart). Run Experiment B identically to Experiment A. Record your data in your tables labeled Experiment B.
When you are finished, quietly tabulate your data from each experiment with your partner and analyze according to the contingency tables described in the flow chart (note there are different contingency tables for each experiment!). Each contingency table answers a different question. Record all work on your data sheets. Have an instructor check your tabulated data and data analysis. Put your results on the board (see flow chart).
Experiment C. Analyze the results on the board for all crickets, using a goodness of fit test. (Read over Example II in the appendix for how to run a chi-square goodness of fit test). This experiment is testing the predictability of cricket behavior. Is cricket behavior is predictable – either always acting the same, or always acting differently when confronted with a different cricket, or is cricket behavior unpredictable? That is, will some crickets act the same when confronted with another cricket and while others will always act differently when confronted with another cricket?
11
FLOWCHART:
PART II
Experiment A Experiment B
After recording data for 15 min, tabulate
your data with your partner to answer:
Did crickets 1 and 2 behave differently?
Solitary
Aggressive
Neutral
Cricket 1
Cricket 2
Experiment C
Did all of the crickets in lab behave the same way in both trials? Can we make any generalizations about cricket behavior?
Perform chi-square goodness-of-fit test on class data.
Did your cricket behave differently in Expt A vs. Expt. B?
Perform chi-square contingency test (directions in Appendix I)
Record your outcome on board
Perform chi-square contingency test
(directions Appendix I)
After completing Expt 2, tabulate your data for your cricket (across both experiments) as follows:
Your cricket
Solitary
Aggressive
Neutral
Expt A
Expt B
Move with your cricket to another group for Expt. B so that it is facing a new cricket.
12
Cricket Behavior Lab Data Sheets Names (yours)_________________________
Experiment A: (your lab partner’s) _____________________
What are you testing?
What is your null hypothesis?
If you reject your null hypothesis, what do you conclude?
DATA FOR EXPERIMENT A
Your Cricket (color _______)
Solitary
Aggressive
Neutral
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
13
Cricket Behavior Lab Data Sheets
DATA FOR EXPERIMENT A
Your Cricket (color _______)
Solitary
Aggressive
Neutral
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
TOTAL:
EXPT A
Your partner’s cricket totals
(color _______)
Solitary
Aggressive
Neutral
TOTAL:
EXPT A
14
Cricket Behavior Lab Data Sheets
OBSERVED DATA FOR EXPERIMENT A
DATA
Solitary
Aggressive
Neutral
ROW TOTAL
Your Cricket
(color _____)
Your partner’s cricket
(color _____)
COL TOTAL
GRAND TL
EXPECTED DATA FOR EXPERIMENT A
Calculate your expected data in each field by multiplying the row total recorded above by the column total and dividing by the grand total.
Solitary
Aggressive
Neutral
Your Cricket
(color _____)
=row tl*col tl
grand tl
Your partner’s cricket
(color _____)
STOP! Examine your expected data. Are any values less than five? If so, you have too few observations to be confident about their validity. Omit those categories in both your expected and observed data tables, calculate new row, column and grand totals, and recalculate your expected data. If you only end up with one column of data, calculate your expected data and degrees of freedom using the goodness of fit test.
Chi-square = sum of all (obs-exp)2
exp
Chi-sq = _________
Df = ____________ (# of rows-1) x (# of columns -1)
Compare your calculated Chi-sq value to the Chi-square table. In what column is your value equal to or greater than the chi-square value in the table? Follow that column up to find the P-value (the probability that your results could be due to random chance). Note: the bigger your chi-square value, the smaller your P-value!
What is your P-value?
_____greater than 0.05?
or
_____ 0.05 or less?
Do you support or reject your null hypothesis for Experiment A?
What does this tell you about your crickets’ behavior in Experiment A?
15
Cricket Behavior Lab Data Sheets
Experiment B:
What are you testing?
What is your null hypothesis?
If you reject your null hypothesis, what do you conclude?
DATA FOR EXPERIMENT B
Your cricket (color _______) with a new cricket
Solitary
Aggressive
Neutral
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
16
Cricket Behavior Lab Data Sheets
DATA FOR EXPERIMENT B
Your cricket (color _______) with a new cricket
Solitary
Aggressive
Neutral
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
TOTAL:
EXPT B
17
Cricket Behavior Lab Data Sheets
OBSERVED DATA FOR EXPERIMENT B
Cricket data
Solitary
Aggressive
Neutral
ROW TOTAL
Your cricket in Expt A
Your cricket in Expt B
COL TOTAL
GRAND TL
EXPECTED DATA FOR EXPERIMENT B
Calculate your expected data in each field by multiplying the row total recorded above by the column total and dividing by the grand total.
Solitary
Aggressive
Neutral
Your cricket in Expt A
=row tl*col tl
grand tl
Your cricket in Expt B
STOP! Examine your expected data. Are any values less than five? If so, you have too few observations to be confident about their validity. Omit those categories in both your expected and observed data tables, calculate new row, column and grand totals, and recalculate your expected data. If you only end up with one column of data, calculate your expected data and degrees of freedom using the goodness of fit test.
Chi-square = sum of all (obs-exp)2
exp
Chi-sq = _________
Df = ____________ (# of rows-1) x (# of columns -1)
Compare your calculated Chi-sq value to the Chi-square table. In what column is your value equal to or greater than the chi-square value in the table? Follow that column up to find the P-value (the probability that your results could be due to random chance). Note: the bigger your chi-square value, the smaller your P-value!
What is your P-value?
_____greater than 0.05?
or
_____ 0.05 or less?
Do you support or reject your null hypothesis for Experiment B?
What does this tell you about your cricket’s behavior in Experiment B?
18
Cricket Behavior Lab Data Sheets
EXPERIMENT C – USING CLASS DATA
What are you testing?
What is your null hypothesis?
If you reject your null hypothesis, what do you conclude?
OBSERVED DATA FOR EXPERIMENT C
SAME (number of groups that accept Null Hypothesis in Exp. B)
DIFFERENT (number of groups that reject Null Hypothesis in Exp. B)
GRAND TOTAL
EXPECTED DATA FOR EXPERIMENT C (Use the GOODNESS OF FIT TEST, not the contingency table for these calculations.)
Calculate the expected data by dividing the grand total by the number of categories, in this case – grand total/2. You would expect each category to have the same number.
SAME (number of groups you would expect to accept Null Hypothesis in Exp. B)
DIFFERENT (number of groups you would expect to reject Null Hypothesis in Exp. B)
Chi-Square calculations:
Chi-square = sum of all (obs-exp)2
exp
Chi-sq = _________
Df = ____________ (# of categories -1)
Compare your calculated Chi-sq value to the Chi-square table. In what column is your value equal to or greater than the chi-square value in the table? Follow that column up to find the P-value (the probability that your results could be due to random chance). Note: the bigger your chi-square value, the smaller your P-value!
What is your P-value?
_____greater than 0.05?
or
_____ 0.05 or less?
Do you support or reject your null hypothesis for Experiment C?
What does this tell you about cricket behavior in Experiment C?
What Students Are Saying About Us
.......... Customer ID: 12*** | Rating: ⭐⭐⭐⭐⭐"Honestly, I was afraid to send my paper to you, but splendidwritings.com proved they are a trustworthy service. My essay was done in less than a day, and I received a brilliant piece. I didn’t even believe it was my essay at first 🙂 Great job, thank you!"
.......... Customer ID: 14***| Rating: ⭐⭐⭐⭐⭐
"The company has some nice prices and good content. I ordered a term paper here and got a very good one. I'll keep ordering from this website."
