
European starlings just beat parrots at their own game – copying complex sounds. In a new study by researchers in Amsterdam and Leiden, the two bird groups were tested on their ability to imitate R2-D2’s famous beeps and whistles from Star Wars.
Across more than 2,000 imitations, the starlings proved superior at reproducing the trickiest tones.
The key difference came down to anatomy. Starlings excelled at duplicating sounds that stack two pitches at once, while parrots matched them only on simpler, single-pitch beeps.
The findings reveal how much vocal skill depends on the physical design of the syrinx – the sound-producing organ deep inside a bird’s throat.
The work was led by Nick C. P. Dam at Leiden University. His research focuses on vocal imitation and comparative vocal learning.
R2-D2’s voice blends a human synthesizer with robotic inflection, providing a repeatable set of beeps and chords. That fixed palette allowed multiple species to tackle identical targets.
The monophonic sounds (single pitch) test how well a bird can copy one tone cleanly, while the multiphonic sounds contain two pitch streams at once, forcing a bird to track overlapping patterns.
Owners and bird lovers supplied video clips through a citizen science program, and researchers annotated each imitation. They worked with 103 birds spanning nine parrot species and European starlings.
Parrots matched starlings when imitating single-pitch units, but the starling advantage returned on two-pitch tasks. Within parrots, budgerigars and cockatiels outperformed African greys and amazons on single-pitch copying.
Multiphonic sounds proved harder for everyone than monophonic ones. Some model units were easier to mimic than others, depending on their pitch complexity and overlap.
Researchers confirmed that several starlings produced two independent pitches at once during their best copies, while parrots never produced two pitches simultaneously within the same unit.
The syrinx, a bird’s voice box at the windpipe split, shapes what sounds are possible. In songbirds it contains two sound sources that can run independently, which enables clean two-pitch output.
That independence underlies biphonation, the ability to produce two pitches at once. Parrots lack separate control of twin oscillators, so they cannot fully match the mirrored contours in the sci-fi sounds.
“Differences in imitation accuracy likely stem from syrinx anatomy rather than perceptual or cognitive disparities.” said Dam. That conclusion does not deny learning; it simply sets boundaries on what learning can reach.
The data revealed a twist that surprises many keepers who admire large talkers. Budgerigars and cockatiels – both small parrots – copied single-pitch units more precisely than African greys and amazons.
Research on parrot brain organization suggests two nested circuits: a core and a surrounding shell linked to vocal learning. The new results hint at a tradeoff, where larger repertoires may carry slight penalties in per-sound precision.
That tradeoff would make sense if small parrots practice fewer imitations more intensely. Larger parrots might stockpile more sounds but polish each one less.
None of this downgrades the remarkable speech feats of African greys. It simply shows that test design can reveal different strengths.
To compare species fairly, the team used dynamic time warping – a method that aligns two time series to measure similarity. They tracked features like pitch over time and matched each imitation to its intended model unit.
Lower dissimilarity scores indicated a closer match. Those scores were then analyzed across species and sound types.
The approach is a workhorse in birdsong analysis because it respects timing differences. It also handles subtle pitch slides that rigid, point-by-point comparisons would miss.
Comparative scoring was done at both the element and unit levels, which prevented rare outlier chirps from skewing the overall results.
Much of the dataset came from pet owners contributing videos from home. That diversity is hard to create with lab birds alone.
The project recruited singing parrots worldwide to test flexible pitch behavior. It shows how carefully curated home recordings can support serious science.
Citizen partners also help test stimuli that birds hear outside the lab. That matters when the model is a cultural icon rather than a natural call.
Home environments vary, so the team could not measure each bird’s training time. Even so, species-level patterns remained consistent.
Hardware can set sharp boundaries on what learners can achieve. Starlings win at two-pitch copying because their instrument can run two voices at once.
Parrots match starlings on single-pitch work because both groups can track one line well. Training and motivation still matter, but neither can add a second oscillator.
Species differences within parrots suggest multiple winning strategies. Some species keep smaller but cleaner sets of imitations, while others stretch for size.
Future work can connect brain circuit size, practice history, and imitation accuracy. It can also test whether accuracy predicts mate choice or social bonding in different species.
The study is published in the journal Scientific Reports.
—–
Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates.
Check us out on EarthSnap, a free app brought to you by Eric Ralls and Earth.com.
—–
