Good scientists do their best to remove bias from their observations and experiments. Statistical methods such as blocking, double-blind trials, use of appropriate controls, and randomization are among their key tools. Results are suspect when hypotheses are designed after data are collected, when confirmation bias, cultural bias, and other errors of judgment creep into experimentation, analysis, and interpretation.

But what type of bias is indicated when scientists study more blue flowers than green? Tall ones rather than short? How does unconscious bias – predilections for favourites – affect our ecological examinations? As Martino et al. state in their May 2021 Nature communication, “Plant scientists’ research attention is skewed towards colourful, conspicuous and broadly distributed flowers”. Charismatic organisms get our attention.

The researchers performed a deep dive into literature associated with studies of plants in the southwestern Alps, determining which species were most researched by scientists exploring the plants in this biodiverse area. They looked at a few hundred published papers and demonstrated that

morphological and colour traits, as well as range size, have significantly more impact on species choice for wild flowering plants than traits related to ecology and rarity.

Specifically, the Martino team found disproportionate research was conducted on pretty flowers.

This analysis indicated how the choice of investigated species across the literature in the last 45 years has been strongly influenced by plant traits related to aesthetics.

Aesthetics. Prettiness. It’s sad that this equates to greater scientific curiosity. But it does.

The researchers found a significant relationship between the number of published papers and flower colour. Blue flowers were the most studied. White and red/pink were much more studied than ugly-duckling brown or green flowers.

. . . traits such as bright colours, accessible inflorescences and conspicuousness are shown to drive research attention, highlight what we call an aesthetic bias in plant research. While aesthetics is today used to refer to art and beauty (often in direct opposition to scientific values like objectivity), the Greek root of the word refers to sensory perception. . .
Here it is interesting to note that humans have evolved trichromacy, that is the separate perception of wavelength ranges corresponding to blue, red and green regions through specialized structures. It has been speculated that the evolutionary acquisition of colour vision in humans and other primates led to an increased ability to locate ripe fruits against a green background. The human eye is thus optimized to perceive green, red and blue which, according to colour psychology theory, also greatly impacts people’s affection, cognition and behaviour. The evolved and physiological aspect of human perception is also demonstrably affected by sociocultural factors, since education, class, gender, age, cultural background all shape how we perceive the world. . . What matters is that this bias affects the representativity of data used to ground research priorities and conservation policies and, as such, risks compromising efforts to effectively focus plant conservation activities and preserve plant biodiversity.

There was also a significant positive effect of plant stem height and a weak negative effect of flower size on research interest – taller plants have more easily accessible inflorescences. (Who wants to crawl on hands and knees to examine ugly flowers?) Finally, the scientists found that there is a positive effect of range size on research interest. This might mean that a broad range makes a species available to more researchers, increasing the likely that it would be studied. But broadly distributed plants are less prone to extinction. For many ecosystems, uncharismatic species with a constricted range might manifest an unexplored, out-sized environmental niche effect of great importance. But we won’t know if we avoid studying ugly organisms.

All of this should be very concerning for anyone interested in understanding ecological systems as completely as possible – without ugly data missing. The authors suggest that “when plant scientists select to study a specific wild plant among the pool of species available in a given study region, it may be that factors unrelated to the biological question end up influencing species choice and introducing biases in the research outcome.” I doubt that many of us recognize how this unintended bias could skew our understanding of an ecological system or landscape.

But I’ll conclude with this contradictory message. If we realize that our research has a bias, and we focus on ways to mitigate it, then maybe it’s not so bad to disproportionately investigate the charismatic. Most of us don’t explore nature in a social vacuum. We are attracted to pretty things. We need funding. We want public appreciation for our conclusions. It’s easier to get money and attention for a project studying the handsome and beguiling than the homely and beleaguered.

Few creatures are as charismatic as bumble bees – except, perhaps honey bees. To me. They have my attention. So, I study them, day and night. But I’ll try to be more aware of the hundreds of other (less charismatic) animals with whom they share their patch of earth.