Undirected singing rate as a non-invasive tool for welfare monitoring in isolated male zebra finches

Abstract

Research on the songbird zebra finch (Taeniopygia guttata) has advanced our behavioral, hormonal, neuronal, and genetic understanding of vocal learning. However, little is known about the impact of typical experimental manipulations on the welfare of these birds. Here we explore whether the undirected singing rate can be used as an indicator of welfare. We tested this idea by performing a post hoc analysis of singing behavior in isolated male zebra finches subjected to interactive white noise, to surgery, or to tethering. We find that the latter two experimental manipulations transiently but reliably decreased singing rates. By contraposition, we infer that a high-sustained singing rate is suggestive of successful coping or improved welfare in these experiments. Our analysis across more than 300 days of song data suggests that a singing rate above a threshold of several hundred song motifs per day implies an absence of an acute stressor or a successful coping with stress. Because singing rate can be measured in a completely automatic fashion, its observation can help to reduce experimenter bias in welfare monitoring. Because singing rate measurements are non-invasive, we expect this study to contribute to the refinement of the current welfare monitoring tools in zebra finches.

Fig 1. Upon isolation, the singing rate gradually increases and remains high thereafter.

(A) The number of daily song motifs produced (colored dots) since the onset of social isolation on day 0. The colored dots show data from individual birds (N = 6) and the lines indicate the trend (see Methods). The dotted vertical line marks the end of the habituation period (see Results) and the dashed vertical lines delimit days 11–13 from which data was taken for panel B. The daily number of motifs produced after long-term isolation (≥ 29 days) remained high (right panel, the duration of the isolation is indicated in the legend). The horizontal lines indicate the number of motifs averaged across three consecutive days. On day zero, only recordings from after the isolation were analyzed (left panel). (B) Bar plot showing the mean motif rates across birds on days 11–13 (left) and at the end of the isolation (right). Individual lines correspond to individual birds (color matching A). (C) Histogram of the daily number of song motifs produced across 19 birds held in isolation for at least 14 days including habituation (N = 315 days after habituation). The minimum motif rate was 629 (see the zoom-in inset, note different x-axis). (D) Histogram of the minimum daily number of song motifs in those 19 birds. The minimum in each animal was computed across the 9+ days it spent in isolation after the habituation period (17±10 days, range 9–46 days).

Fig 2. The singing rate is not affected by interactive WN.

(A) Spectrograms of a song motif that did not trigger (black box) and that triggered a short burst of WN broadcasted through a loudspeaker (WN, red box). The vertical white dashed lines mark the onsets and offsets of the targeted syllable. (B) The number of motifs in each bird (colored lines) as a function of time since the onset of the interactive WN. The bold black line shows the average across birds before WN onset and the bold red line the average after WN onset (N = 26 birds, where for four birds there is partial data before WN onset). WN was triggered whenever the pitch of the targeted syllable was below (or above) a manually set threshold. (C) Example of a bird that compressed the syllable duration to trigger more noise. Syllable renditions triggering WN are marked in red, all others in black. The vertical rectangles mark the days from which the data in A were taken from. (D) Bar plot of the average daily song motif counts during days without WN (left bar) and with WN (right bar) for N = 26 birds. Lines indicate average daily song motifs in individual birds. (E) Histogram of the number of song motifs produced across all days from all N = 43 birds (including birds without manipulation, Fig 1C), where days without any manipulation are indicated in black (793 days from N = 43 birds) and days with WN are shown in dark gray (294 days from N = 26 birds).

Fig 3. Brain surgeries transiently decrease the singing rate.

The number of daily song motifs produced from five days before the surgery up to 10 days after the surgery (N = 12). Surgeries consisted of either an injection of saline (green, N = 3) or ibotenic acid (blue and orange, N = 9) into a brain area outside the known song-control system. On day zero, only recordings from after the surgery were analyzed. Different line styles indicate different birds and the black line indicates the average over all birds.

Fig 4. Tethering transiently decreases the singing rate.

(A) The number of song motifs produced in the 24 hours before tethering onset (left bar) and the 24 hours after tethering onset (right bar, lines connect individual birds). Birds were either tethered before surgery but after weight attachment (N = 4, weight attachment), after microdrive implantation surgery (N = 5, microdrive) or after stimulation electrode implantation surgery (N = 8, stim implant). Some birds (N = 3, cage transfer) were also transferred into a smaller cage on the same day. All birds were tethered for the first time except the bird that sang around 2000 motif renditions after tethering (in A, stim implant), this bird is marked with a green arrow in panel B where singing rate from the first time tethering 1 day after surgery is shown (see Methods). (B) The number of song motifs increased gradually after the start of tethering on day 0 (day 0 corresponds to the first 24 hours after tethering). The number of days between the surgery and tethering onset is shown for each animal as a (blue or green) number, where green marks tethered-only birds, blue marks birds that also changed the cage on the same day, and black marks the average over all birds. The bird that was tethered 10 days after surgery has been tethered before. Five birds received an LED implant for optogenetics, the circles mark the days with optogenetic manipulations. All other birds received an implant for electrical stimulation. The diverse line styles indicate the corresponding birds in Panel A; solid lines mark birds without data available 24 hours before tethering.

Fig 5. A welfare threshold on singing rate and the posterior probability of manipulation.

(A) The proportion of birds not having fully recovered from the manipulation as a function of time since onset of the manipulation for interactive WN (WN, blue, N = 26, Fig 2B), surgery (yellow, N = 12, Fig 3), and tethering (purple, N = 15, Fig 4E), i.e., the proportion of birds with a singing rate below the welfare threshold W₉₉. For illustration, also shown are birds subjected to isolation (dashed red, N = 6, Fig 1A). (B) Posterior probability of a recent manipulation (here tethering), as a function of the singing rate.