Sunflowers and light: 'optical illusions' may steer how plants grow
· BehindLines newsroom
A team at Tel Aviv University proposes that plants may experience the equivalent of "optical illusions": the direction in which a young stem bends would not always match exactly the direction from which the most light comes. The study was published online in the Proceedings of the National Academy of Sciences (PNAS) on 22 September 2026; Phys.org carried the university's press release on 2 November. The work is older, however: a preliminary version, not yet peer-reviewed, has been on the bioRxiv server since around the beginning of April 2026.
Computing without a brain
According to the team of Aharon Kempinski, Amir Porat, Mathieu Rivière and Professor Yasmine Meroz, light receptors spread around the stem each sense the local light. Each signal is transformed non-linearly, and these signals are then added as vectors to give the growth direction. No central processing is involved. Yasmine Meroz sums up, as quoted by phys.org, that plants can integrate complex information about their environment without a brain or a central system, the computation emerging from the way light is sensed locally along the stem.
What was measured
In the published article, the researchers worked on sunflower seedlings (Helianthus annuus, variety HAEMEK-6) lit by blue diodes of 450 nanometres. Phys.org specifies that their movements were filmed for more than twenty hours. With a single source, the final tilt angle increased with intensity, between roughly 0.7 and 7 watts per square metre. With two opposing sources, the response varied non-proportionally to the difference in intensity. The 0.73 coefficient of determination concerns the single-source fit.
The most telling test places two opposing sources of almost identical intensity (about 3.85 W/m²) and a third, weaker one (about 1.74 W/m²) perpendicular to them. The directional influences of the two strong sources cancel and the seedlings grow toward the weak source, which the authors interpret as evidence that the plant does not simply seek the maximum of physical light.
The "illusion" remains a prediction
This is where caution is needed. The model predicts a maximum gap of about 6 degrees between the most brightly lit direction and the direction actually followed. According to the published article, however, that gap is below the resolution of the experiment: the natural variability of growth (standard error of about 12 degrees) masks it. Phys.org stresses this too: these distortions could not be measured directly. To be clear, speaking of an optical illusion does not mean that plants see images.
Other limits stated by the authors: only one species and one wavelength were tested, with about 16 to 36 seedlings per condition. The theoretical framework is three-dimensional, with planar reductions for the one- and two-source tests. The idea put forward by Yasmine Meroz that this principle extends to other cells or to bacteria is a hypothesis, not a tested result.
The full text of the final version is available in PMC.