Mike Ko Personal Portfolio

 

Home-School Education
2001-2012
Hong Kong

University of Durham
Bachelor of Science
2014-2017
United Kingdom

University of Sussex

Master of Arts
2017-2018
United Kingdom




Academic Works
                                Year 1
: Organisms & Environment

Investigation of the Significance of Rate and Duration of Display Behaviour in Determining Winners in Betta splendens

 

Introduction

            Mating behaviours are an integral part of animal life, as they are often the means to achieve an individual’s goal of reproducing. Decisions made by individuals during mating also directly influence their offspring’s characteristics and thus the evolutionary path of the species. These are partly shaped by the selective pressures that act on the species. Apart from showing the dynamics of animal mating, such behaviours can illustrate a species’ evolution and how the environment might have influenced it.  

            One aspect of mating behaviours is intra-sexual male competition within a given species. Males of certain species must often engage in mutual confrontations for access to mates, which can be quite costly. As such, aggressive display behaviours have variously evolved to relay levels of physical prowess and other traits between participants (Enquist et al., 1988). This communication can potentially resolve confrontations before actual fighting occurs. The Siamese fighting fish, Betta splendens, is one species of which males are known to engage in display behaviour. The species mainly occur in the lower Mekong river of Southeast Asia (Rainboth, 1996). Male members of the species will exhibit ‘facing’ and ‘broadside’ displays against each other when competing for mates.

            Although mating displays can potentially avert physical combat, they can still incur energy or metabolic costs (Castro et al., 2006). Consequently, competing males might adopt certain strategies to minimise such costs. For example, males may choose to regulate their rate and duration of displays in ways that are most cost-effective. This choice of strategy can then highlight the crucial factors that determine the dominant ‘winner’ of display contests. Accordingly, this study examined the rate and duration of displays of B. splendens to assess which is the more significant factor in yielding winners.


Method

            The basic setup for the experiment involved a small fish tank with the approximate dimensions of 20 x 20 x 40 cm. An opaque and an adjacent clear glass partition separate the tank into two roughly equal sides. One side contains a heater tube at the far left corner. Two vertical lines spaced approximately half the length of B. splendens (3.25 cm) away from the central partitions were labelled on the tank. One B. splendens individual of a single colour was placed in each side of the tank. Figure 1 illustrates the basic setup.

Figure 1. Fish Tank Basic Experimental Setup

 

            The first part of the experiment involved measurement of the rate and duration of facing displays for fishes on both sides of the tank. Facing displays were characterised by the flaring of the gill operculum and exposing the branchiostegal membrane whilst the fish directly faces its opponent. A mirror was used to elicit display behaviour from each fish, upon which the cumulative total time and number of displays over 5 minutes were recorded.

The second part involved the determination of the winning and losing fish in the pair. The opaque barrier at the centre of the tank was removed to allow both fishes to engage in display behaviour against one another. The total time in which each fish was halfway within the zone between the glass partition and the vertical line label over five minutes was then recorded. The fish that spent a longer time in the defined zone was determined as the winner.

            Water temperature was around 24 °C on the side without the heater. The experimental procedures described above were repeated for separate fish pairs to give 100 total replicates. However, only 89 of the 100 replicates were included in the raw data. The remaining 11 replicates had winners that won by £5s and were thus judged to be statistically insignificant. In addition, there was one outlier data pair for mean rate of display (replicate 22), which was appears to be a due to a data-handling mistake. It was thus excluded from the statistical analysis and figures to avoid a slight skewering of the data.

            To evaluate whether there was a significant difference between the rate and duration of display between winner and losers, the two-sample (unpaired) two-tailed t-test was used. The test was used with the assumption that the relevant data were normally distributed, as the sample size was deemed to be sufficiently large (i.e. <40 sample pairs).


Results

            Table 1 shows the raw data obtained from the experiment, excluding one replicate fish pair. Table 2 includes certain figures derived from the raw data. The mean values for rate of display between winners and losers are very close (8.45 s and 8.17 s, respectively), as illustrated by Figure 2a. In contrast, the mean duration of displays varied more substantially (92.85 s for winners and 71.95 for losers), a trend shown by Figure 2b. Figures 3a and 3b shows the distribution of the above data for winners and losers.

            Statistical analysis did not find any evidence that the mean rate of display were different between winners and losers (|t|: 0.373; tcrit: 1.966; P: 0.71; df: 407). In contrast, the same test found some evidence that winners had different mean display duration relative to losers (|t|: 2.490; tcrit: 1.984; P: 0.01; df: 101).

            Winners on average spent much more time in the defined zone than losers when both fishes displayed against each other (mean time in zone: 269.6 s and 185.4 s, respectively).

REPLICATE

Winner Rate of Display

Winner Duration of Display (s)

Winner Time in Zone (s)

Loser Rate of Display

Loser Duration of Display (s)

Loser Time in Zone (s)

1

6.8

56.6

297

5.2

37.2

67

2

6.8

72

33

3.8

16.7

14.6

3

1.4

4.8

194.4

10.4

102.9

43.6

4

10.8

34

278

9

137

72

5

1

16

286.1

6.3

55.2

255.8

6

4.6

83.9

294.7

3.2

112.4

258.8

7

4.4

138

223

5.2

33

44

8

9

42.5

295

4

41.2

262

9

9.4

120

284

4.8

79

272

10

5

115

288

7.2

80

242

11

12.6

152

288

5.2

125

207

12

8.8

134

279

6.2

75.9

203.6

13

4.8

44

274

13

63

146

14

1.2

233

299

5.8

226

290

15

1.6

194

282

5.2

51

128

16

9.4

121

298

3

48

197

17

3.8

95.8

278.6

6.4

69

51.8

18

4.4

97

278

4.2

47

265

19

7.6

240

289

2.4

14

194

20

9.4

184

289

7

29

260

21

12.6

84

291

3.6

17

113

22

63

89

295

64

80

278

23

8

110.4

220.7

3.6

22.8

21.2

24

5

117

127

3.8

176

38

25

5.2

120

296

6.6

94

287

26

3.4

14.7

289

5.8

32.5

57

27

22

10.7

386

28

3.5

306

28

6

30

65

3

2.2

21

29

8

142

288

3.4

39

266

30

9.2

204

290

6.2

95

246

31

36

41

276

28

76

82

32

6.8

164

293

8

175

268

33

11

47

286

6.2

17

190

34

6.4

100

273.3

11.2

180

265.3

35

10.2

112

287

11.6

81

188

36

10.4

94

300

3.4

35

237

37

4.8

128

281

5.4

101

114

38

7

154

261

8.6

98

18

39

4

119

267

5.2

51

117

40

6.4

170.3

300

5.8

157.5

292.3

41

15.4

122.2

300

8

57.2

245.9

42

1.2

2.6

179

4.7

19.3

26.4

43

11.6

84

300

4.8

28.5

259

44

10.4

120.3

286.9

11

91.3

281.3

45

2.6

38.2

160.5

2

0

43.3

46

1

5.1

61.1

4.4

29.5

25.8

47

3.4

11.8

286.8

8

149.1

132.2

48

2.6

21.2

274

8

21

123

49

8.8

68

281

5.6

46

219

50

4.6

30

279

12.2

19.2

264

51

5.6

164

179

3.4

17

42

52

4.5

26.6

294

11.4

43.4

288

53

13

136

283

2.4

5

271

54

8.4

47.3

286.5

13

118.9

268.8

55

8

56

290

11.4

143

266

56

2.4

41.5

300

5.8

42.5

120

57

9.2

110

296

11

118

269

58

18.4

130.5

300

14.4

108.8

270

59

13

69.8

276

5.4

34.7

247

60

5

66.5

292.8

8.8

158.9

263.1

61

4.4

144

259

9.4

62

194

62

10.2

58.6

296

2.8

24.5

147

63

4.6

190

289

11

100

138

64

20

30

283

12

14

265

65

13.4

120.7

294.8

6.4

46.8

168.1

66

4.6

85

264

5

21

60

67

10.2

134

295

4.2

82

263

68

2.6

84

282

3.6

15

97

69

15.2

93

287

17.8

68

200

70

4.8

51

287

8.6

62

276

71

7.2

51.4

298

2.8

3.8

275

72

15.2

56

285

28.2

121

274

73

1

14.6

230.4

2

0

180.1

74

3.6

27.1

267.8

9.4

75.1

201.7

75

2.6

103

278

2.2

1

227

76

6.8

78

267.6

6.2

93

198

77

3.4

71

192

2.6

5

31

78

5.2

124

289

5

269

258

79

8.8

124

293

13.2

130

243

80

10.2

97.6

297.2

11

96.8

88.6

81

10.6

81

292.4

10

103

278.5

82

15.2

115

287

9

151

236

83

4.6

44.6

300

7

141

278.8

84

8.8

112.1

293.1

10.4

71.4

215.4

85

13.8

72

300

4

37

189

86

2.8

236

245

9.2

86

58

87

3.2

60

298

4.8

46

292

88

15.2

65

272

6.8

39

202.8

89

6.8

161

295

8

212

251

Table 1. Raw Data on the Rate of Display, Duration of Display and Time Spent in Defined Zone for Winner and Loser B. splendens.

 

Winner

Loser

 

Rate of Display

Duration of Display (s)

Time in Zone (s)

Rate of Display

Duration of Display (s)

Time in Zone (s)

Mean

8.45

92.85

269.92

8.17

71.95

186.40

Standard Deviation

7.96

55.79

53.79

7.84

56.32

91.14

First Quartile

4.40

47.30

-

4.20

29.00

-

Median

6.80

89.00

-

6.20

59.60

-

Third Quartile

10.20

124.00

-

9.40

101.00

-

Minimum

1.00

2.60

-

2.00

0.00

-

Maximum

36.00

240.00

-

28.20

269.00

-

Table 2. Selected Statistical Values for the Rate of Display, Duration of Display and Time in Zone for Winner and Loser B. splendens

(a) (b)

Figure 2. a) Mean Rate of Facing Display and b) Mean Display Duration for Winning and Losing B. splendens

 

(a) (b)
              

Figure 3. (a) Boxplot of Rate of Display Data Distribution for Winning and Losing B. splendens;
               (b) Boxplot of Duration of Display Data Distribution for Winning and Losing B. splendens

 

Discussion

            The study found no significant difference in the rate of facing displays and a statistically significant difference in duration of displays between winning and losing B. splendens males.

            Overall, the findings seem to indicate that the duration of the facing displays is the significant proximate factor in determining winners. The lack of difference in rate of display potentially alludes to a selective pressure against high frequency of display. This could be rationalised by the fact that displays incur energy costs. Castro et al. (2006) has shown that male B. splendens winners exhibited greater oxygen consumption at night. As such, overly frequent displaying behaviour may lead to fatigue, reducing energy available for other activities. This would apply to both winners and losers, hence it can be speculated a universal selective pressure that constrains frequent displaying is at work. Although prolonging displays would also require energy, it directly increases the chance of winning a display contest. Assuming that the energy expended in not significantly greater, increasing display duration could thus ultimately contribute to an increase in fitness.

            Of course, duration of display is a proximate factor that is is ultimately determined by other variables. These can include perceived strength, age and size of an opponent as well as the surrounding landscape. Possible routes for further investigation might include the assessment of specific relationships between such variables and display duration.

            This study’s experimental design has several limitations. In particular, it has been observed that fishes respond slightly differently against mirror images relative to conspecifics. Although they still confront mirror images with the same kind of response, they appear to show an elevated level of fear as well (Desjardins and Fernald, 2010). Depending on whether fear causes lowered or heightened aggression, this study’s data may either be more modest or exaggerated than normal.

            In addition, male display contests do not occur in isolation in nature. The presence of other fishes as an audience can alter the display behaviour of the contest participants. Indeed, contesting males gave more intense displays around females and milder ones around male audiences (Doutrelant, McGregor and Oliveira 2000). Variations in the response to such conspecifics in individuals could affect the relevance of this study’s data and findings.

            It should also be noted that conditions in the experiment’s fish tank were asymmetrical, namely that one side has a submerged heating tube. The possible relationship between temperature and display duration has not been tracked in this study. Indeed, it is possible that display duration may have a positive correlation with water temperature. The presence of a nearby object may also represent a potential refuge, thus encouraging fish contestants to concede and retreat. Hence the presence of the heater in one side of the tank may serve as a confounding factor in this study.

            Finally, the recording of cumulative display times and (to a lesser degree) frequency is somewhat dependent on the observer’s reaction time. Although the actual discrepancies may be small, there was inevitably greater room for human error.


References

 

Back to Top