1. Ecology
  2. Neuroscience

Drift in individual behavioral phenotype as a strategy for unpredictable worlds

  1. Ryan T Maloney  Is a corresponding author
  2. Athena Q Ye
  3. Sam-Keny Saint-Pre
  4. Tom Alisch
  5. David M Zimmerman
  6. Nicole C Pittoors
  7. Benjamin L de Bivort
  1. Department of Organismic and Evolutionary Biology, Harvard University, United States
  2. Department of Psychology and Neuroscience, Colorado College, United States
  3. Tufts University, United States
  4. Department of Physics, Harvard University, United States
https://doi.org/10.7554/eLife.103585.3
Share this article
Cite this article
  1. Ryan T Maloney
  2. Athena Q Ye
  3. Sam-Keny Saint-Pre
  4. Tom Alisch
  5. David M Zimmerman
  6. Nicole C Pittoors
  7. Benjamin L de Bivort
(2026)
Drift in individual behavioral phenotype as a strategy for unpredictable worlds
eLife 13:RP103585.
https://doi.org/10.7554/eLife.103585.3
  2. Download BibTeX
  3. Download .RIS
4 figures, 3 tables and 1 additional file

Figures

Figure 1 with 1 supplement
Download asset Open asset
Characterizing changes in individual preferences in Drosophila melanogaster.

(A) 2 hr sample of centroid-tracking data for a fly in a circular arena. Each point is colored based on whether it is moving CCW, CW, or radially in the arena. (B) Sample of 100 hr of continuous recording for 4 individual flies (colors). Hourly means of turning indices are shown in light lines. Dashed lines are low-pass filtered with a timescale cutoff of 24 hr. Solid lines are low-pass filtered shuffled data showing the average tendency across the experiment. (C) Mean power spectrum of turning data for all continuously monitored flies (n=252) for actual and shuffled data. Shaded areas represent 95% confidence intervals generated via bootstrapping (n=1000). (D) Schematic of Y-maze assay. Flies make either a left or right turn each time they walk through the intersection. (E) Standard deviation of daily right biases vs average right bias across days for individual flies (points). Colors indicate DGRP genotype (n=48–235 for each genotype (see Table 1)). (F) Mean individual handedness per day across all DGRP lines, a random subset are thickened to show representative changes in individual handedness over time. (G) Autoregressive model of individual right bias over time with parameters to estimate the initial right bias variability (σBet-Hedging) and rate of daily change in right bias (σDrift). (H) Posterior estimates of σDrift, σBet-Hedging, and ϕ (the autoregressive parameter characterizing the rate of reversion to zero bias), for each DGRP genotype. Grey bars represent 95% credible intervals. Lines indicate the smallest q value between populations. (I) As in H, posterior estimates of right bias variability parameters for flies treated with 5-HTP, AMW, and controls, as well as mutant flies with a missense mutation in trh generated by in vivo CRISPR. n=98 or 192 for each condition (see Table 1). q values <.05 are indicated.

Figure 1—figure supplement 1
Download asset Open asset
Change in behavioral biases over time and the influence of genetics and serotonin on rate of change.

(A) Statistics of the ‘circling’ behavior metric. A1: Example of one fly’s centroid position over time, colored by circling (where -1 equals CCW and 1 equals CW). A2: Histogram of circling measurements across all frames and all flies. A3: Average power spectral density across all flies as estimated via Lomb-Scargle. Shaded regions indicate bootstrapped 95% confidence intervals (n=1000 replicates). B1-3: As A, except for speed (pixels/s, each pixel = .28 mm). C1-3: As A-B, except measuring of heading velocity (change in direction between frames of the flies [change in heading angle between timepoints]). D1-3: As A-C, except measuring the normalized distance of the fly from the center of the arena, in units of unit diameter (28 mm) (E) Bivariate posterior estimates of σBet−hedging, and φ for each DGRP line. Concentric circles represent deciles of posterior density, with the outer circle representing 95% posterior density. (F) As E, except for experiments with manipulation of serotonin with 5-HTP, AMW, and trhn. (G) Posterior estimates of distribution of differences between batches in hierarchical model for bet-hedging, drift, and φ (see Figure 1 for modell). (I) Estimates of 1-day R value for all pairs of sequential experiments by treatment condition. (J) Estimates of distribution of possible R by condition, as estimated by bootstrapping. (K-M) As (I-J), except for experiments with trhn mutants. p values in J,M Fisher Z-Transform.

Figure 2 with 1 supplement
Download asset Open asset
Modeling adaptive scenarios of phenotypic drift.

(A) In a simplified model in which both phenotypes and environments have two states, the optimal fraction of the population that should change preference (fshift) over a period of time equals the probability the environment changes (pshift). See Appendix 1. (B) Model for the number of individuals with a particular continuous preference as a function of time in a fluctuating environment and individual age. nϕ,α,t, is the number of individuals with preference ϕ, age α at day t in the simulation. Two different cases determine this value, one for flies surviving from the previous timestep (α>0), and one for flies born in a particular timestep (α=0). The number of flies of a particular behavioral phenotype surviving on each successive day is determined by a function of how far that preference is from the ideal preference on that day (orange), the total number of flies that already have, or drift into having that phenotype on that day (red), and a bounding term (purple) that stops the distribution of preferences from diffusing away from the general range of what is adaptive (adding some degree of reversion to the mean consistent with our observed values of ϕ, and via the Wiener-Khinchin theorem, consistent with the power spectrum observed in Figure 1C). The key behavioral strategy parameter from these terms is σd, which determines the rate at which flies’ preferences drift over time. The number of new flies born each day is given by the total number of flies above the age of reproductive maturity amin (red) times the birth rate β. New flies are born with an initial preference from a normal distribution centered on the long-term environmental mean with a standard deviation given by σbet-hedging (blue) (C) Example environmental fluctuations and corresponding fitness landscape showing the change in population over 100 simulation days for differing amounts of phenotypic drift and bet-hedging. Green dot indicates ideal strategy (ii) (D) Population over time for strategies marked with Roman numerals in (C). (E, F) As in (C) for two additional example environmental fluctuation patterns.

Figure 2—figure supplement 1
Download asset Open asset
Distribution of animal preferences over time for different strategies.

Strategies from Figure 2C-D, with the fraction of the starting population (colormap) per preference bin (Y axis) plotted over time (x-axis). Mean preferences are reflected in the central preference (as can be noted in cases with no bet-hedging [i, iv, None]), as newly born flies begin with the mean preference. The total population summing across all preferences produces the total population graph in 2D.

Figure 3 with 1 supplement
Download asset Open asset
Effects of amplitude of environmental fluctuation, frequency of fluctuation, and age of reproductive maturity on ideal amounts of bet-hedging and phenotypic drift.

(A) Fitness landscapes over combinations of environmental fluctuation amplitude and frequency. Each heatmap shows the geometric mean of the log change in population for each combination of drift and bet-hedging over 100 randomized environments. Heatmaps are normalized to their maximum and minimum values. Rows of heatmaps have the same environmental fluctuation frequency, and columns have the same environmental fluctuation amplitude (as measured by the standard deviation of all timepoints σMean). The nine amplitude and frequency combinations in this panel correspond to values denoted with white boxes in (C). (B) As (A), except columns of heat maps have the same age of reproductive maturity, as determined by amin. (C) Optimal amounts of bet-hedging (warm color scale; left) and drift (cool color scale; right) for each combination of environmental fluctuation amplitude and frequency. White squares indicate values associated with heatmaps in A. Dotted blue line corresponds to an environmental fluctuation period of 20 days, which is twice the age of reproductive maturity in these simulations. (D) As (C), except for optimal amounts of bet-hedging and drift for each combination of environmental fluctuation frequency and amin. Dotted blue line corresponds to environmental fluctuations of twice the age of reproductive maturity. White squares indicate values associated with heatmaps in (B).

Figure 3—figure supplement 1
Download asset Open asset
Effect of birthrate and longer timescales on strategies.

(A) Adjusting birthrate β (rows) changes the overall rate of change of the population (left column) without affecting the ideal amount of bet-hedging or drift (middle and right columns). (B) As Figure 3D, but with logarithmic scale for Fluctuation Period and Age of reproductive maturity.

Figure 4 with 1 supplement
Download asset Open asset
Optimal bet-hedging and phenotypic drift strategies for real-world environmental fluctuations.

(A) 1000 days of relative humidity data from the Arikaree River in Colorado, USA (left panels) were used to generate an environmental selection filter with low (top-left) or high (bottom-left) amplitude fluctuations (σmean in Figure 2B). Fitness landscape heatmaps over bet-hedging and drift strategies for different ages of reproductive maturity (amin). (B) As in (A), using average daily temperature data from Longreach, Australia. (C) Pipeline for comparing the optimal variability strategies of organisms subject to real-world environmental fluctuations. Daily environmental time series from many sites were collected, normalized, and used in the model to produce fitness landscapes over σBet-Hedging and σDrift. All landscapes were then subject to principal components analysis. These simulations held amin and σmean constant. See Methods. The loadings of PC1 (97.8% of the variance; top-right) indicate that this component encodes the optimal amount of bet-hedging, while PC2 (1.9% of the variance; bottom-right) encodes optimal drift. (D) Environmental time series from specific locations, plotted on PC2 vs PC1 axes, colored by optimal amount of bet-hedging. (E) As in (D), except color indicates optimal amount of drift. (F) As in (D), except color indicates the type of environmental measurement. (G) As in (D), except color indicates the Köppen climate classification of their location.

Figure 4—figure supplement 1
Download asset Open asset
Breakdown of simulations by Koppen climate classification and measurement type.

(A) Map of all stations used for the simulation, colored by broad categories of Koppen climate classification. Subcategories indicated with colors for B. (B) Mean Ideal Drift and Bet-hedging as a function of age of reproductive maturity and scaling of environmental amplitudes, averaged across all time series of a given Koppen climate classification type. (C) Average Power Spectral density of each environmental measurement type. Colored dots correspond to data in (D). (D) As (B), but averaged across each type of measurement.

Tables

Table 1
Drosophila melanogaster genotypes used in this paper.
GenotypeSourceFigurenCitation
Canton-SBDSC 643491 A-C, S1A-D250 (24hr) 252 (2hr)
DGRP 45BDSC 281281E, 1 G, S1E55Mackay et al., 2012
DGRP 85BDSC 282741E, 1 G, S1E22Mackay et al., 2012
DGRP 105BDSC 281391E, 1 G, S1E91Mackay et al., 2012
DGRP 208BDSC 251741E, 1 G, S1E111Mackay et al., 2012
DGRP 426BDSC 281961E, 1 G, S1E235Mackay et al., 2012
DGRP 535BDSC 282081E, 1 G, S1E115Mackay et al., 2012
DGRP 703BDSC 282181E, 1 G, S1E48Mackay et al., 2012
DGRP 796BDSC 282331E, 1 G, S1E140Mackay et al., 2012
DGRP 819BDSC 282421E, 1 G, S1E145Mackay et al., 2012
DGRP 907BDSC 282621E, 1 G, S1E141Mackay et al., 2012
Isod1 (Oregon R)Clandinin Lab1H-I S(H-M)192 each (1 H: AMW, 5HTP, Control), 98 (1I: Control)Silies et al., 2013
trhnThis study1I, S1 G,K-M98
GS01997BDSC 91886
Act-Cas9BDSC 54590
Table 2
Model parameters.
Figureet (Environment series)et scaling timeσeσDσBσmaxaminβ (Birth Rate)Sim. length (days)
Figure 2C-2, Figure 3—figure supplement 1Temporally filtered white noise0.30.1250–30–331040100
Figure 2ETemporally filtered white noise0.30.1250–30–331040100
Figure 2FTemporally filtered white noise0.30.1250–30–331040100
Figure 3A,CTemporally filtered white noise0–30.1250–0.050–0.01310401001
Figure 3B,DTemporally filtered white noise0.30.1250–0.050–0.0132–72401001
Figure 4ARH, Arikarree River (NEON)0.1, 0.30.1250–0.10–0.1310, 60, 360401001
Figure 4BAvg. Temp, Longreach AU (NOAA)0.1, 0.30.1250–0.10–0.1310, 60, 360401001
Figure 4C-G, Figure 3—figure supplement 1ANOAA/NEON0.20.1250–0.10–0.5310401001
Figure 3—figure supplement 1B-CTemporally filtered white Noise0.30.1250–0.10–0.132–1024401001
Figure 4—figure supplement 1B,DNOAA/NEON0.1, 0.2, 0.30.1250–30–3310, 60, 360401001
Table 3
List of NEON datasets used.
Product ID / DOI / data product nameSite IDsDates / date accessed
DP1.20217.001
https://doi.org/10.48443/br51-rd19
Temperature of groundwater		WLOU, WALK, TOOK, TOMB, SYCA, SUGG, REDB, PRPO, PRLA, PRIN, POSE, OKSR, MCDI, MAYF, MART, LIRO, LEWI, KING, HOPB, GUIL, FLNT, CRAM, COMO, CARI, BLWA, BLUE, BLDE, BIGC, BARC, ARIK2016-03-04 to 2022-04-30 DA: 06/28/2022
DP1.00024.001
https://doi.org/10.48443/51ss-fm81
Photosynthetically active radiation (PAR)		WLOU, WALK, TOOK, TOMB, TECR, SYCA, SUGG, REDB, PRPO, PRLA, PRIN, POSE, OKSR, MCRA, MCDI, MAYF, MART, LIRO, LEWI, LECO, KING, HOPB, GUIL, FLNT, CUPE, CRAM, COMO, CARI, BLWA, BLUE, BLDE, BIGC, BARC, ARIK2016-03-04 to 2022-03-31 DA: 05/04/2022
DP1.20261.001
https://doi.org/10.48443/jnwy-xy08
Photosynthetically active radiation below water surface		TOOK, TOMB, SUGG, PRPO, PRLA, LIRO, FLNT, CRAM, BLWA, BARC2017-07-28 to 2022-03-31 DA: 05/04/2022
DP1.00004.001
https://doi.org/10.48443/rt4v-kz04
Barometric pressure		YELL, WREF, WOOD, WALK, UNDE, UKFS, TREE, TOOL, TOOK, TECR, TEAK, TALL, SYCA, SUGG, STER, STEI, SRER, SOAP, SJER, SERC, SCBI, RMNP, REDB, PUUM, PRPO, PRLA, PRIN, POSE, OSBS, ORNL, ONAQ, OKSR, OAES, NOGP, NIWO, MOAB, MLBS, MCRA, MCDI, MAYF, MART, LIRO, LEWI, LENO, LECO, LAJA, KONZ, KONA, KING, JORN, JERC, HOPB, HEAL, HARV, GUIL, GUAN, GRSM, FLNT, DSNY, DELA, DEJU, DCFS, CUPE, CRAM, CPER, COMO, CLBJ, CARI, BONA, BLUE, BLDE, BLAN, BIGC, BART, BARR, BARC, ARIK, ABBY2013-09-12 to 2022-01-31 DA: 03/13/2022
DP1.00098.001
https://doi.org/10.48443/k9vk-5k27
Relative humidity		ABBY, ARIK, BARC, BARR, BART, BIGC, BLAN, BLDE, BLUE, BONA, CARI, CLBJ, COMO, CPER, CRAM, CUPE, DCFS, DEJU, DELA, DSNY, FLNT, GRSM, GUAN, GUIL, HARV, HEAL, HOPB, JERC, JORN, KING, KONA, KONZ, LAJA, LECO, LENO, LEWI, LIRO, MART, MAYF, MCDI, MCRA, MLBS, MOAB, NIWO, NOGP, OAES, OKSR, ONAQ, ORNL, OSBS, POSE, PRIN, PRLA, PRPO, PUUM, REDB, RMNP, SCBI, SERC, SJER, SOAP, SRER, STEI, STER, SUGG, SYCA, TALL, TEAK, TECR, TOOK, TOOL, TREE, UKFS, UNDE, WALK, WLOU, WOOD, WREF, YELL2013-09-12 to 2022-02-28 DA: 03/14/2022
DP1.00023.001
https://doi.org/10.48443/9qpc-5v70
Shortwave and longwave radiation (net radiometer)		BART, UKFS, TALL, YELL, WREF, WOOD, WLOU, WALK, UNDE, TREE, TOOL, TOOK, TECR, TEAK, SYCA, SUGG, STEI, SRER, SOAP, SJER, SERC, SCBI, RMNP, REDB, PUUM, PRPO, PRLA, PRIN, POSE, OSBS, ORNL, ONAQ, OKSR, OAES, NOGP, NIWO, MOAB, MLBS, MCRA, MCDI, MAYF, MART, LIRO, LEWI, LENO, LECO, LAJA, KONZ, KONA, KING, JORN, JERC, HOPB, HEAL, HARV, GUIL, GUAN, GRSM, FLNT, DSNY, DELA, DEJU, DCFS, CUPE, CRAM, CPER, COMO, CLBJ, CARI, BONA, BLUE, BLDE, BLAN, BIGC, BARR, BARC, ARIK, ABBY2013-09-12 to 2022-05-31 DA: 06/07/2022
DP1.00005.001
https://doi.org/10.48443/jqb2-vy96
IR biological temperature		YELL, WREF, WOOD, UNDE, UKFS, TREE, TOOL, TEAK, TALL, STER, STEI, SRER, SOAP, SJER, SERC, SCBI, RMNP, PUUM, OSBS, ORNL, ONAQ, OAES, NOGP, NIWO, MOAB, MLBS, LENO, LAJA, KONZ, KONA, JORN, JERC, HEAL, HARV, GUAN, GRSM, DSNY, DELA, DEJU, DCFS, CPER, CLBJ, BONA, BLAN, BART, BARR2013-09-12 to 2022-01-31 DA: 03/14/2022
DP1.20288.001
https://doi.org/10.48443/t7rj-pk25
Water quality		TOOK, TOMB, SUGG, PRPO, PRLA, LIRO, FLNT, CRAM, BLWA, BARC, ARIK, MART, WALK, TECR, SYCA, REDB, PRIN, POSE, OKSR, MCRA, MCDI, MAYF, WLOU, LEWI, LECO, KING, HOPB, GUIL, CUPE, COMO, CARI, BLUE, BLDE, BIGC2014-01-11 to 2022-03-16 DA: 05/04/2022

Additional files

MDAR checklist
https://cdn.elifesciences.org/articles/103585/elife-103585-mdarchecklist1-v1.docx

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Ryan T Maloney
  2. Athena Q Ye
  3. Sam-Keny Saint-Pre
  4. Tom Alisch
  5. David M Zimmerman
  6. Nicole C Pittoors
  7. Benjamin L de Bivort
(2026)
Drift in individual behavioral phenotype as a strategy for unpredictable worlds
eLife 13:RP103585.
https://doi.org/10.7554/eLife.103585.3