Extreme Variability and Forward Telescoping in University Students' Estimates of Macroevolutionary Deep Time
Laura R. Novick, Psychological Sciences, Vanderbilt University
Some understanding of deep time is necessary for understanding macroevolution, which in turn is critical for understanding and managing Earth's biodiversity. Because most research on deep time has been conducted from a geoscience perspective, Kefyn Catley and I recently took an evolutionary and psychological approach to investigating this topic (Catley and Novick, 2009). We assessed university students' knowledge of deep time with respect to seven key historical and evolutionary events: (a) the origin of the Earth (approximately 4-4.6 billion years ago [bya]), (b) the oldest rocks with evidence of life (i.e., the first fossils; ≈3.5 bya), (c) the appearance of eukaryotic cells (those with a nucleus and organelles; ≈1.75 bya), (d) the "sudden" appearance of a large number of animals comprising many different body plans (i.e., the Cambrian explosion; ≈500-600 million years ago [mya]), (e) the first appearance of mammals (≈200 mya), (f) the extinction of most dinosaurs (≈65 mya), and (g) the appearance of the first hominid, Homo habilis (≈2 mya).
The subjects in our study were 126 students recruited from psychology, education, and biology classes at two universities. Sixteen of these students failed to provide any time estimates. The remaining 110 students, who were the focus of our article, were assigned to stronger and weaker background groups based on their college coursework in biology. Those who had taken at least the first semester of the year-long biology class for majors were assigned to the stronger biology background group (n = 54; M = 3.2 semesters of college biology or relevant geology courses). All other students were assigned to the weaker biology background group (n = 56; M = 0.5 semesters of relevant coursework).
Dating events from Earth's history is a very different task for people who believe the Biblical account of creation than for those who accept modern science. In particular, creationists would be expected to give a very low estimate for the age of the Earth and to exhibit a relatively flat function for the remaining time estimates. Three students provided time estimates that fit this pattern. The results reported in the remainder of this essay are for the remaining 107 students.
We found that students' time estimates spanned a startlingly large range. As noted earlier, the correct answers range from 2 mya to 4 bya. Students' answers, however, ranged from 800 years ago (the time of the Magna Carta, St. Thomas Aquinas, and the mathematician Fibonacci) to 600 billion years ago (586 billion years before the big bang)! Moreover, this extreme variability of the time estimates occurred for each question, not just across the set of questions. Another characteristic of students' time estimates was forward telescoping, meaning that the events were generally estimated to have occurred more recently than they actually did. This phenomenon has been documented in the psychological and consumer survey literatures with respect to people's time estimates for events that occurred in the recent past (within the past 50 years). We provided the first documented evidence of forward telescoping for estimates of deep time, indicating that this is a general characteristic of people's representations of time. Across the seven events, stronger background students (even the 19 who had taken a college evolution class) were hardly more accurate than their weaker background peers, although they did tend to provide estimates that spanned a somewhat smaller range.
As an illustrative example, consider Figure 1, which shows students' time estimates for the extinction of most dinosaurs. Note that the X-axis is on a logarithmic scale, so the actual time frame is much larger than it appears. I have added graphics for key historical events to the X-axis to provide a more concrete perspective on this scale. Although it appears in the histogram that stronger background students were more likely than weaker background students to provide an answer of the correct order of magnitude (marked with a blue arrow), the difference is not statistically significant. Two-thirds of students underestimated the extinction date.
We also examined students' relative time estimates by dividing each student's absolute time estimate for each event by his/her estimate of Earth's age. This allowed us to examine students' knowledge of the spacing of the events and also provided a clearer picture of students' patterns of over- and underestimation. The results of a cluster analysis of these data revealed five groups of students whose relative time estimates followed distinctly different patterns. (The three apparent creationists exhibited a sixth pattern of responding.) Although it is unclear what factors are responsible for the different patterns, we ruled out general knowledge of biology as a contributing factor as each cluster included approximately equal proportions of stronger and weaker background students.
The results of this study suggest that college students generally do not have an effective conceptual framework to make sense of evolutionary deep time. In our 2009 paper, we proposed an instructional strategy that uses a phylogenetic tree to introduce key evolutionary events in terms of their "markers" and spacing, but the efficacy of this strategy remains untested at this time (March, 2012).
Catley, K. M., & Novick, L. R. (2009). Digging deep: Exploring college students' knowledge of macroevolutionary time. Journal of Research in Science Teaching, 46, 311-332.