Acquired predator recognition via epidermal alarm cues but not dietary alarm cues by isolated pupfish
Data files
Sep 13, 2023 version files 14.83 KB
Abstract
We tested whether Shoshone pupfish Cyprinodon nevadensis shoshone and Amargosa River pupfish C. n. amargosae respond behaviourally to conspecific chemical alarm cues released when epidermal tissue is damaged by a predator. We found that both subspecies reduced activity and vertical position in the water column in response to alarm cues. We then tested if pupfish can use alarm cues to acquire recognition of a novel predator. We trained pupfish with (1) water + odour of largemouth bass fed a diet of earthworms, (2) alarm cues from skin extract (epidermal alarm cues) + odour of bass fed a diet of earthworms, or (3) water + odour of bass fed a diet of pupfish (dietary alarm cues). Pupfish responded to epidermal alarm cues but not to dietary alarm cues. Pupfish were retested with the odour of bass that were fed an earthworm diet. Pupfish that had previously received epidermal alarm cues reduced vertical position and activity relative to the other two treatments. This is the first demonstration of acquired recognition of a novel predator by a pupfish, the first report of partial predator naiveté, and opens the possibility of predator-recognition training as a tool for management and conservation of endangered desert fishes.
Description of the data and file structure
There are two data files. The first file WIsenden et bal pupfish alarm reactions contains data from the first experiment testing alarm reactions by Amargosa and Shoshone pupfish to conspecific alarm cues derived from skin extract.
Column headings are:
- Subspecies = which subspecies was being tested, either Shoshone pupfish Cyprinodon nevadensis shoshone, of Amargosa River pupfish, C.n. amargosae
- Am1Sh2 = the treatment code used by the stats package to identify which subspecies was tested
- Treatment = which chemical cue was added to the tank, which was either consubspecific alarm cue derived from skin extract, or water (control)
- Specific treatment = the treatment in text rather than number code
- Pre_Activity = The number of grid lines crossed in the 5-min pre-stimulus observation period
- Post_Activity = The number of grid lines crossed in the 5-min post-stimulus period after the test cue was added
- PreVert_Total = pre-stimulus score of vertical distribution using the grid drawn on the front of the tank. This is total of observations taken every 10 s for 5 min. After revision, we changed the analysis to the average grid position from 1-5.
- PostVert_Total = post-stimulus vertical distribution score, summed over the 5 min. Data reported in the revised ms used the average score.
- Sex: F for female and M for Male
- TL_mm = total length in mm
- Notes: stuff that happened
The second file Wisenden et al. SH Learning contains the data from the second experiment testing if Shoshone pupfish could use alarm cues to acquire recognition of the odor of a novel predator. Behaviors recorded for both experiments were changes in activity and vertical distribution before (PreAct, PreVD) and after (PostAct, PostVD) introduction of test cue. All the other column headings are intuitive, to us, but just in case, here they are:
- Conditioning date = the date mm/dd/yy when the conditioning trial was conducted.
- Conditioning trial = the number assigned to the trial
- Conditioning treatment = the combination of cues presented to pupfish during the trial. The odor of largemouth bass fed a diet of earthworms (to control for diet effects) + alarm cue derived from pupfish skin, bass odor (earthworm diet) + water to control for previous exposure to bass odor, and bass odor on a pupfish diet, which potentially contains dietary alarm cues.
- TreatmentEpi1W2Diet3 = numerical codes for the three treatments in the conditioning trials
- Cond_ ** = values from conditioning trials
- Test_ ** = values from test trials when each fish was retested with bass odor (earthworm diet)
- VD = vertical distribution scored as the horizontal row in the grid occupied by the fish either before (pre) or after (post) exposure to the cue. Note that after revision we switched from using the total scores to using the average scores
- Act = activity scored by the number of grid lines crossed in each 5-min observation period either before (pre) or after (post) exposure to cue
- Sex: either female or male
- Length = total length in cm, converted to mm during analysis
- Notes = stuff that happened. For example, we ran out of Shoshone males and had to complete the expt using Amargosa males, as indicated
Behavioral observation of fish in lab aquaria.
Statistical treatments are described in the manuscript.
Microsoft Excel.