Unpredictability 2.0

By Ethan Young

 

I saw those Djakarta markets for what they were: fragile, precious things. The people who sold their goods there might have been poor, poorer even that the folks in Altgeld. They hauled 50 pounds of firewood on their back every day, they ate little, they died young. And yet for all that poverty, there remained in their lives a discernable order, a tapestry of trading routes and middle men, bribes to pay, and customs to observe, the habits of a generation played out every day. It was the absence of such coherence that made a place like Altgeld so desperate, I thought to myself; it was the loss of order.

-Barack Obama (1995), Dreams from my Father

 

Barack Obama spent part of his childhood living in Jakarta, Indonesia, and his emerging adulthood working as a community organizer in Altgeld Gardens, a public housing project on Chicago’s South Side. In his memoir, Obama described both Jakarta and Altgeld as tough environments in which people died young. Indeed, Jakarta and Altgeld seemed similar on a fundamental dimension of environmental risk—harshness—which Ellis and colleagues (2009) defined as age-specific rates of morbidity-mortality. At the same time, Jakarta and Altgeld appeared very different on a second key dimension of environmental risk: unpredictability. The challenges and struggles of life in Jakarta were tough but similar from day-to-day. In Altgeld, life was chaotic, haphazard, and disordered.

For decades, scholars have examined the link between environmental conditions, evolution, and the development of life-history strategies. In particular, the literature on environmental unpredictability has grown substantially in the past 10 years. However, this body of work faces two major theoretical ambiguities. The first is the definition of environmental unpredictability as a selection pressure across evolutionary time. Specifically, unpredictability is often defined as the level of spatial-temporal variation in environmental harshness. However, this definition does not specify the pattern of variation in statistical terms. Different patterns of variation might produce different selection pressures, resulting in different adaptations.

A second ambiguity concerns the proximate mechanisms—adaptations—that evolved to detect and respond to environmental unpredictability. There are at least two frameworks. The first is the ancestral cue perspective, which proposes that humans evolved to detect and respond to cues that reliably indicated high environmental unpredictability across evolutionary history. The second is the statistical learning perspective, which suggests that organisms estimate the level of unpredictability by integrating differences in lived experiences across development. Drawing on one or the other approach has consequences for hypotheses and measurement.

Our paper refines the definition of unpredictability in two ways. First, we identify the patterns of variation that make environments more or less predictable. Second, we address stationarity, which refers to whether or not the statistical structure of an environment itself changes over a lifetime.

Stationary environments have three main statistical properties: variance, autocorrelation, and cue reliability. Variance refers to average deviations from the mean. For example, high temporal variance in harshness means that the environment can vary widely from safe to dangerous (around a mean value) across time. However, whether or not high variance is unpredictable depends on whether the level of harshness is also autocorrelated. Autocorrelation refers to the degree to which current conditions are related to future conditions. Variation can be predictable if it is autocorrelated, even when variance is high. Cue reliability refers to the extent to which observations provide information (reduce uncertainty) about current or future environmental conditions. Cues may provide information about current or future states of the environment, even if these states are not autocorrelated.

 

First, we identify the patterns of variation that make environments more or less predictable. Second, we address stationarity, which refers to whether or not the statistical structure of an environment itself changes over a lifetime.

 

In non-stationary environments, the statistical structure of the environment changes over time. However, some non-stationary patterns are more predictable than others. For example, predictable non- stationary environments could have a trend (e.g., slope), seasonal variation, and/or cyclic variation. However, unpredictable non-stationary environments might contain random change points, i.e., abrupt changes in statistical properties of the environment. For example, there might be a sudden increase in mean levels of harshness (e.g., a natural disaster) or an increase in the variance of resource distribution (e.g., the stock market crashing).

Having refined definitions of unpredictability, an open question is which proximate mechanisms organisms use to detect environmental unpredictability. We discuss two possibilities. The first is the ancestral cue approach to unpredictability, which is anchored in the general ‘ancestral cue’ perspective in evolutionary psychology. This perspective assumes that ancestral environments included cues that were associated with fitness-relevant changes in environmental conditions. As a consequence, natural selection has favored brains to treat these cues as privileged sources of information.

The second perspective is the statistical learning approach. This approach states that natural selection has shaped proximate mechanisms to track the statistical structure of the environment by integrating differences in lived experiences across development, without privileging particular sources of information per se. Organism use their experiences as raw data to build models of the statistical structure of the environment (not necessarily consciously). They then use these models to ‘estimate’ (i.e., adapt to) such properties as the overall mean level, variance, and autocorrelation in harshness.

The approaches target the same process—estimating environmental unpredictability—but differ in the types of information that trigger a response. Ancestral cues carry information about ancestral environments. If particular cues, for example geographic relocations, were reliably associated with environmental unpredictability, then natural selection may have equipped the mind to detect and respond directly to the cue. In contrast, a statistical learning proximate mechanism responds directly to the statistical patterns of change in its environment. Specifically, the organism responds to environmental unpredictability when it detects a prediction error. For example, if the level of danger does not change after a relocation, the statistical learning mechanism will not make a prediction error, and therefore it will not trigger a response to unpredictability.

Of course, ancestral cue detection and statistical learning are not mutually exclusive. In fact, organisms could leverage both sources of information. For instance, organisms could use lived experience as ‘raw data’ where each experience is weighted equally. If exposed to an ancestral cue, they could add them to their predictive models, using weights to account for the ancestral knowledge contained in ancestral cues. Or, ancestral cues could indicate when the individual should recalibrate their model of the environment. For example, an organism could use lived experience and, upon detection of an ancestral cue, such as a geographic relocation, it may be adaptive to re-estimate these statistics because a move could mean that past conditions are no longer informative for predicting future conditions. Thus, the organism might throw out “old data” in favor of using “new data” after a transition occurs.

However, these mechanisms interact, empirical studies drawing on the ancestral cue perspective need to select and measure the cues that are hypothesized to have indicated environmental unpredictability in our evolutionary past. In contrast, studies drawing on the statistical learning perspective need to measure lived experiences across time and compile enough observations to model patterns of variation (e.g., variance, autocorrelation, change points etc.). To do so, researchers could use time series data and analytical techniques for characterizing patterns of change over time.

Future research should collect measures derived from both approaches and integrate insights. Doing so will refine our understanding of environmental unpredictability and its connection to life history development.

 

 

Read the paper: Theory and measurement of environmental unpredictability