Nathanet al.,Science 375 , eabg1780 (2022) 18 February 2022 8 of 12
hedge grassland other
Land use classDisplacement (m/s)Relative selectionstrength1.02.02.0 2.5wheaturban wheat hedge grassland other1.5ABFStudy DesignData CollectionData ProcessingData AnalysisD35.560 35.62033.11033.050 0 2 km500
50
5FixesRoost9.320 9.36056.37556.3900 1 km2.721 2.723 2.72539.46539.46739.4690 100 mRaw
Filtered
CTCRW
HydrophoneG HAnimal cognition experimentsTime in open (s)^0250 500(^84)
(^04)
(^02)
0
24
0
Time swimming (s)
0
200
400
Mean
dispersion (m)
Speed (m*h
-1)
Lab Field
Activity
Boldness
ID 26 ID 34
34
34 34
2634 26
26 26
Lab Field
ID 26
Unrealistic movement
Filtered data
Median smooth (K = 11)
53.245
53.246
5.258 5.260 5.262
0 50 m
I
E
0
1
0.5
(^10)
. 0
(^1).
0
(^5).
0
0 6 12 18 24
0 6 12 18 24
Time of day
Overall dynamicbody acceleration
Stationary state
probabilities
Cruising/gliding Active swimming
High-energetic movement
Straightness index
1
(^6).
0
(^8).
0
0.4
Non-
shortcuts
ShortcutsTrans.
returns
After trans.
returns
35.500 35.740
33.0000 5 km
33.240
35.500 35.740
33.000
33.240
0 5 km
64.505
64.504
19.043 19.045
0 50 m
C
Hydrophone
Tracking
radar
Computer
vision
Cellular
tower
Server
System
receiver
ICARUS GNSS
Ground
station
ARGOS
Reverse-GPS
GPS logger
GPS w/
remote upload
Triangulation
Reverse-GPS
GPS w/
remote upload
Solar
geolocators
Doppler shift
Time of arrival
Direction of arrival
Signal strength
Data link
Localization signal
PSAT
Data exploration Filtering & smoothing Filtering & modeling
Kernel density estimation Step selection function Hidden Markov modeling
Animal personality experiments
RESEARCH | REVIEW