Religion and Evolutionary Psychology

Alice Berry Spring 2012

Religion, Evolutionary Psychology and Human Development;

A Survey of Current Theory.


My husband, who is a lawyer, once had a colleague on a case

who was a Mormon. The case was in Utah, so the other lawyer was

on home turf, and felt comfortable talking about his faith to a

visitor. He was apparently an excellent litigator, very smart,

and so my husband was all the more surprised when he told of the

myth of the seagull, a foundational story in Mormonism. He

described the previously rational lawyer as getting a softer look

in his eyes, an expression of wonder on his face, and fascination

with the story of how seagulls came to the middle of the desert

and saved the followers of Joseph Smith, the holy man leading a

holy cause.

What is it about people, humans, that makes them, makes us,

so receptive to belief in events or beings that exist outside of

our empirical perception? There is an attitude within certain

sections of the scientific community that the population of

believers is less rational and are less open-minded than those

2


who profess to be above such anti-rational thought, and in some

circles, believers are considered delusional (Dawkins, 2006).

Evolutionary Psychology (EP) theorists believe they have arrived

at an answer. A combination of evolutional biology and cognitive

psychology, EP proposes evolutionary reasons for the pervasive

existence of religion across all societies.

I Modules and Evolution

Arguably a universal aspect of human development, there is

no society that has been found to have developed without some

form of worship of a deity or designation of sacredness to an

object or person. Previous religious theory held that religion

was a central concept, with all of its attributes and behaviors

emanating from a formally held idea, sometimes against a

background of a particular culture (Hick, 2004). Cognitive

scientists have come to embrace a theory known as the “Modular

Model” of human thought, delegating specific processing tasks to

discrete domains in the brain called modules, (Fodor, 1983). This

mental architecture has evolved according to the rules of natural

selection, with modular adaptations that promote the survival of

the organism being carried forward to the next generation.

3


Beginning with just a few modules, the theory has grown to

encompass hundreds if not thousands of “mental rooms,” which are

the places where certain systems operate, and they lead to a

central conceptual processor.

Each of these mental modules, physically or functionally

insulated subsystems, affects some aspect of religious behavior,

belief and influence or direct human behavior and development in

a variety of different spheres, personal and public. What is now

shown in EP research is that there are apparently a whole host of

modules that are especially receptive to counter-intuitive

belief. There are differing sides to these discussions, but

according to the faction that believes religion is an adaptation

in and of itself, certain modules exist to foster mental systems

needed to create religious belief and its behaviors as a survival

mechanism. Using the body of information of traditional religious

scholarship to test evolutionary hypotheses, they reject the

“spandrel, or by-product theory. They say that quantitative

methods refine but do not define scientific inquiry (Wilson, D. S.

2005) and suggest that religion is a necessary system in and of

itself. The larger faction, which includes several high profile

4


writers and theorists, believe that religion is a by-product of a

variety of adaptations, or exaptations (Gould, 1991). These

theorists believe that all the behaviors and personal concepts

that form religion are systems that were originally developed or

evolved to perform some other function. Systems mentioned are

overactive agency detection and Theory of Mind, from which arise

counterintuitive concepts and beliefs. Taboos, sacredness, and

the performance of rituals for protection and purity arise from

the contagion avoidance system. Modules for empathy and emotion

develop beliefs that Gods are interested parties in our moral

judgment, have moral empathy, and engage in social exchange.

(Boyer, 2003)

II. Religious Constructs Evolved Step by Step.

The concept behind EP of religion relies on a set of evolved

cognitive steps the human mind takes to arrive at religious

belief, practice and community. This paper will look at five

conceptual theories with studies and experiments to illustrate

these steps. The (1) Agency Detection Device (ADD) (2) leads to

a theory of mind (ToM) being assigned to an agent. This agent,

who is elusive and unseen, (3) becomes a counterintuitive

5


concept, which is then (4) anthropomorphized . Finally the

concept becomes a construct, (5) which binds people into groups

of believers, or moral communities.

1. Agency

Our brains have been wired to notice any unexplained

movement or action, and this very first step is the granting of

agency to either animate or inanimate objects. A rustle in the

bushes could be a tiger, a rockfall could be a bear. The

responses of this brain module, called the Agency Detection

Device, are also used in a variety of ways to explain the

movement of objects for which there is no empirical explanation.

The module makes a leap of inference to imagine some agent behind

the motion. The imposition of agency on an object (real or not)

also grants it intention.

Justin Barrett and Amanda Hankes Johnson (2003) studied the

assignment of agency to objects. They note the most famous work

in this area, the geometric representations of triangles and

squares of Heider (1944), which demonstrated that moving, visual

displays of triangles, discs and other geometric shapes readily

evoke attributions of belief, desire, emotional states and even

6


gender. They also cite developmental research showing that

infants and children sometimes do attribute “social” causation to

geometric shapes, contending that the forms are self propelled

and goal directed, (Gergely, Nadasdy, Csibra & Biro, 1995;

Rochat, Morgan & Carpenter 1997).

Given that its activation is so hypersensitive, when does

the ADD turn off? The survival advantages would be quickly

nullified if we were never able to disengage the ADD when the

event or object in question is not an actual agent. Barrett and

Johnson’s study challenges the claim that attribution of agency

is only a perceptual process, and that control over objects’

movements mitigates attribution. Their hypothesis for the study

is that people are more inclined to attribute agency when the

object appears to move (1) in an apparently non-inertial (e.g.

self propelled), goal directed path, and (2) independent of the

control of a human agent. The null hypothesis is that control

over the movement is irrelevant, and agency attribution arises

simply through the self propelled-ness and goal directed-ness of

the object. Their study extends the research into the realm of

three dimensional objects, with the authors arguing that if video

7


displays (considered to be perceived through a modular process)

are sufficient for evoking attributions of agency, a three

dimensional display should evoke the same sorts of attributions.

The method of the experiment was: 31 undergraduates, 17

males and 14 females who were randomly assigned to either the In-

Control condition or the No-Control condition. A board with 30

divots was placed on top of a box containing hidden

electromagnets. Each was powered by a switch under the table. The

In-Control condition had the box displaying two small lights and

two switches. The experimenter controlled the lights using the

same hidden switch box as the electromagnets. In the No-Control

condition, no lights were used. Other materials included two tape

recorders, one set of headphones and a blindfolded teddy bear.

The main task of the experiment was to fill the divots of

the puzzle board with ball bearings, and to explain to the teddy

bear what was happening while the subjects were trying to

complete the task. (They had been primed for this with warm-up

exercises). The In-Control subjects were instructed to flip the

switch on the right to the “on” position when the right light

came on, when it went off, they were to turn it off. The No-

8


Control subjects received no instructions regarding the lights

and switches. While the participants were playing, the

experimenter activated the magnet (in the No-Control condition)

or the light (in the Control condition). In both conditions the

magnets came on at comparable intervals, but in the control

condition, the participants activated the magnets, and in the No-

Control condition, the experimenter activated the magnets. When

the magnets were turned on, the balls launched themselves from

their holes, colliding with each other.

The results were derived from transcriptions of the

descriptions directed towards the bear, and were of language only

used for animals or persons, not for language only appropriate

for marbles, lights or switches. 12 of the 31 participants made

some sort of agent attributions to the marbles, 9 (60% of the 15)

in the No-Control and 3 (18.8% of the 16) in the control group.

The mean number of attributions for the no control group was 1.53

(SD=1.96) as compared to .56 (SD=1.75) for the In-Control group,

(although almost all of the attributions in the In-Control group

were made by 1 person, calling the bear Mister and naming a

9


marble Bob). Without this outlier, the Control group averaged .13

(SD=.35) attributions each.

A majority of participants in the no control condition

attributed agency to the metal balls as in previous studies with

animated displays, although without the richness or consistency

evoked by animation. Transcripts revealed that in both conditions

participants were aware that magnets were at work, but only

attributed agency when the subjects had no control over the

movements of the marbles. The researchers question whether the

natural context (actual objects rather than animated displays)

created the difference between their results and the previous

motion picture displays, thinking conceptual factors of the

displays contribute to attribution of agency and could be the

subject of further research. The authors conclude that when an

objects’ action violates our own sense of causal efficacy, we

attribute agency to them.

2. It’s Real!

In addition to agency, we also apply Theory of Mind (ToM) to

unexplained phenomena, once we’ve decided it has intention. The

author Jesse Bering (2004) states that “the capacity to search

10


for reasons for having certain life experiences is religion’s

most important psychological feature. The capacity to attribute

psychological causes to otherwise random events...may be a

distinctly human specialization”.

Bering’s work in the development of religious thought in

children serves to illustrate his points about how we assign ToM

to random action, beginning in childhood. He cites a growing

corpus of research, suggesting that children are very receptive

to religious beliefs, so much so that they seem prepared to do so

(Kelemen, 2004). Other research in this area has found that

“children do not represent God as just another sort of Person,

but rather understand God as a different sort of agent with some

non-human properties (Barrett 2001).

Where do children get the idea of supernatural agents to

begin with? Bering realizes this is a perfectly confounded

question, as there is no child in any culture who has not heard

some stories or references to some cultural religious concept, or

more specifically, God. Bering sees the development of afterlife

beliefs as a place to begin seeking, an area where children

11


appear to spontaneously generate inferences about the continued

existence of consciousness after death.

In an experiment with children between the ages of 4 and 12,

Bering (&Bjorklund, 2004) presented a puppet show with an

anthropomorphized mouse. The mouse was introduced in the context

of a story including information about the mouse’s current

psychological states, including epistemic (the mouse is thinking

about his mother), desire, (the mouse wants to go home),

emotional, (the mouse is sad that he can’t find his way home),

perceptual, (the mouse can hear the birds singing), and

psychobiological, (the mouse is thirsty).

After the children heard this story, an alligator crept out

of the bushes and ate the mouse. Children were then asked a

standardized, counterbalanced series of questions about the

psychological status of the dead mouse. (“Now that the mouse is

not alive anymore, is he still thinking about his

mother?, ...want to go home?, etc.) Responses to such queries

were classified as either continuity reasoning, (the child

reasoned that the capacity for the state in question continued

after death) or discontinuity reasoning (The child reasoned that

12


the specific capacity for the psychological state in question

discontinued after death).

Results showed a developmental trend, with preschoolers

reasoning that dead agents possess a full complement of mental

abilities after death. As they age, children are more likely to

say that the psychobiological states are extinguished, while

emotional, epistemic, and desire states continue. Even

preschoolers understand that biological function stops after

death, but they do not tie these functions to their attendant

states until grade school; the same child who

clearly understands dead mice don’t drink water, will also tell

you very soberly that the mouse is still thirsty.

In a modified study with adults, even people who classified

themselves as “non-believers” found it difficult to deny a dead

human character the capacity to experience emotions, think, have

knowledge, or have desires (Bering & Bjorklund, 2004). We imbue

the dead with our own minds, but because we have experienced the

absence of perceptual and psychobiological states (we know or can

imagine what it feels like to blinded and hungry), we are better

able to reason that dead agents lack these “low level”

13


psychological capacities. Bering suggests that the simulation

constraint be borne out by additional data, but he notes that if

children’s cognitive systems are naturally equipped to breed

ghosts because of simulation constraints, this may help set the

stage for supernatural attribution. There were no figures given

in this study summary as to the number of participants involved,

so the power is not known.

In another experiment, Bering and Baumann (2004)

investigated the cognitive developmental mechanisms that underlie

a child’s ability to represent random events as “standing for”

the intentions of a supernatural agent. Children between the age

of 3 and 7 were asked to play a hiding game guessing the location

of a ball in one of two boxes. Children were awarded prizes

(stickers) if they guesses the correct box. In reality, there

were 2 balls, and the “correct” guess was a function of the hand

movement of the child and which box they chose. Once children

understood the basic task demands, they were shown a picture of a

friendly, make believe, (i.e. two-dimensional cartoon) female

character. The children were read a script that explained that

“Princess Alice” (PA) is magical, and she can become invisible,

14


telling the child that the princess really likes them and that

she will help them play the game. The children were told the

princess will tell them when they pick the wrong box, although

they are not told how they will know but somehow, Princess Alice

will tell them. Children were administered four hiding game

trials, two of which were identical to training trials where

nothing out of the ordinary happened. Regardless of the location

of the child’s hand, they were awarded the prize. On the other

two trials, the unexpected happened (from the child’s point of

view anyway). Either the lamp flashed on and off in quick

succession, (an ambiguous random event caused by the

experimenter) or the portrait fell off the door to the floor (an

iconic random event). Children received the prize on these trials

only if they moved their hand and kept it there in response to

the random event.

After the trials, children were asked to provide verbal

judgments about the causes of these unexpected occurrences. The

children’s responses again reflected a strong developmental

pattern. Only 16% of the 3 and 4 year olds moved their hand to

choose the correct box in response to the event, and only 31% of

15


the 5 to 6 year olds did. By age 7, nearly all the children (82%)

changed their choice by moving their hand in response to the

random event. This developmental trend was also present in the

post-experiment verbal judgments as to why the light flashed or

the portrait fell. The younger children provided mostly non-

agentive responses, (“I don’t know”, or “Because the picture

didn’t stick very well”). Only 18% mentioned Princess Alice at

all, and only 6% thought she was trying to share information with

them.

The 5 and 6 year olds thought PA was an intentional agent,

with 46% of them saying she caused the event, but they did not

know why she did so, or that she “just wanted to”. Only 19% of

the children in this group claimed that PA was trying to help

them find the ball. Among the 7 year olds, the most frequent

response was a “declarative agentive”, in which the child stated

that PA caused the event and was trying to help them find the

ball. 65% of the oldest children in the study gave these verbal

judgments.

In another series of ongoing experiments with the PA

character, children are introduced to her and told she is

16


invisible and that “ She likes good boys and girls, I bet she

really likes you” and are shown a box with a prize that is “very

special and just for them”. They are told that they will be able

to guess what’s in the box and get a prize, but before they do,

the experimenter has to go out to talk to their parent for a few

moments. They are then reminded of PA, and are told, “ Don’t

worry, you won’t be alone, PA will be here with you.” The

children are observed by video monitor, and if they make an

effort to open the box and “cheat”, the second experimenter will

make a table lamp flash on and off. After this, the child is

observed for another 30 seconds to see whether they will continue

looking in the box despite the unexpected event (a continuative

response), or withdraw their hand from the box (an inhibitive

response). Although all the data has not been collected for the

between subjects control condition, (the same experiment is

conducted, but the children hear nothing of PA), preliminary

evidence suggests even the youngest children have an inhibitive

response after encountering the unexpected event in the midst of

their cheating. They looked as they had been caught red handed,

some looked dejected and fearful, and many commented on the event

17


when the experimenter returned to the room. One little girl

exclaimed “Princess Alice is Real!”.

The author notes that developmental research is beginning to

find clues as to whether the cognitive architecture underlying

causal religious belief is an adaptive system whose component

parts converged through selective processes. He also acknowledges

that the evolutionary scenario is speculative and tentative, but

he sees a need to account for the strong influence supernatural

causal beliefs have on human behavior if we are to understand

religion as a natural and human phenomenon.

3a. Counterintuitive Concepts

After assigning agency, and intuiting understanding of the

minds of those agents, the next link in the chain of religious

belief development is the counterintuitive agent, the arrival of

the God concept that brings us to religious belief, practice and

community. Pascal Boyer (2003) posits that a combination of

ordinary and extraordinary properties are within the structure of

counterintuitive concepts. Those ideas and concepts which become

or describe counterintuitive agents can fall into one or a

combination of the following categories: attention arresting,

18


having inference potential, being difficult or impossible to

process, or emotionally provocative. Each of these selection

criteria depends on evolutionarily poised and prepared

architecture in the brain. There are two principal ways to

render a counterintuitive concept, first by violating intuitive

expectations, or by transferring concepts from one ontological

category to another. Applying either a violation or a transfer to

any of the five ontological categories of human thought (animal,

plant, person, natural object or artifact), we get a set of 30

supernatural templates, each according to whether the violation

or transfer is of a biological, psychological or social nature,

with psychological ones the most commonly found (people tend to

project states of mind more than anything else) (Atran, 2002).

The ontological violations or commitments that kick in at this

stage of processing produce the supernatural concepts.

Verification is not actually needed, but only the reasonable

possibility of such verification (someone who is heard but not

seen is visible in principle, but an invisible being is not)

(Barrett, 1996).

19


“Supernatural agent concepts tend to be more emotionally

powerful because they trigger evolutionary survival templates.

This also makes them attention arresting and more memorable.” So

says the archeologist Scott Atran in his book, “In Gods We Trust”

(2002). In a study regarding ways in which ideas become

memorable, Atran, along with Ara Norenzayan (2002), test the

concepts of empirically evident, or intuitive (INT) and minimally

counterintuitive (MCI) beliefs.

Participants were 107 undergraduate students, who were told

they were in an experiment about memory and were given a list of

items to remember without a story context. MCI beliefs were

generated by transferring a property from it’s intuitive domain

to a novel domain (e.g. thirsty door, closing cat). For each MCI

belief, there was a corresponding INT belief, (thirsty cat,

closing door). Thus each word,- “cat”, “door”, “closing”, and

“thirsty”- was equally likely to appear in an INT item as in an

MCI item. This resulted in a set of four statements that achieved

a counterbalanced design, each word in each statement serving as

its own control.

20


Recall was measured in two ways: planned free recall after a

three minute delay, and a surprise free recall after a one week

delay. The latter measure was considered the more important one,

better reflecting the role of recall in oral traditions. A

complex pattern of recall emerged for INT and MCI beliefs.

Contrary to recent studies by Barrett and Nyhof (2001) which

showed participants remembering 92% of MCI items but only 71% of

INT items, Atran’s study showed better recall rates for INT

beliefs than for MCI beliefs. The only difference was when

counter-intuitives made up the majority of beliefs, in which case

there were no differences in recall rates. Because of matching

the two kinds of beliefs the authors conclude that that it was

the intuitive factor that contributed to the recall advantage of

the INT beliefs.

The authors conclude that mostly intuitive belief sets

combined with a few minimally counterintuitive ones had the

highest rate of delayed recall and the lowest rate of memory

degradation over time. They assert that this is the recipe for

successful transmission of cultural beliefs, and it is the

cognitive template that characterizes most popular folktales and

21


religious narratives. A small proportion of MCI beliefs give

advantages over stories with none or too many, just as a

moderately spiced dish is more palatable that a bland one or one

that is too spicy. The right combination of intuitive belief with

just a bit of magic makes for a religious concept that is just

right, and becomes eminently transmissable.

3b. Counterintuitive Constructs

The needed basic assumptions for understanding discourse

about God are supplied by ignoring the ontological distance. The

idea proposed by Atran (Norenzayan, 2002) and Boyer (2003) that

the right amount of violations to certain ontological categories

are necessary to make a Counterintuitive concept stick becomes

problematic when confronted with the Abrahamaic God. Since this

God occupies a different ontological category, it is unclear what

intuitions apply to God’s ontology either by theologians or

common believers.

Contrary to Boyer’s ontological categories, Justin Barrett

stresses the centrality of anthropomorphism (Guthrie 1993), as a

foundational cognitive bias. Despite theological descriptions,

people seem to incorporate anthropomorphism and naturalistic

22


characterizations into their intuitive God concepts, contrary to

theological tenets. Barrett, writing in his paper

“Conceptualizing a Nonnatural Entity: Anthropomorphism in God

Concepts” (1996) presents the God concepts of three major

monotheistic religions, Catholicism, Judaisim, and Islam, (Keil,

1996) and speaks of similarities in the theological descriptions

of God in each. In all three, God is infinite, limitless, all

perfect, all powerful, unchanging, non-material, all knowing, and

perfectly simple. If these religions, which are the basis of

Western theological concepts, attribute to God a vastly different

type of existence than our own, how do we cross the ontological

gap and understand God? The implication is that the Abrahamic God

is outside of the ontological category of “Person”.

There is a difference between the theological description of

a religious concept, or God, and how God is spoken of and related

to “on the ground”. God is often spoken of by non-theologians in

very naturalistic terms, as if God is super-human. Sometimes the

descriptive talk is blatantly anthropomorphic as in “the hand of

God” and other times more subtle as in “God sees...”. Another

subtle use places God in a temporal context, as in “then

23


God...” . Prefacing a phrase about God with “then” places an

atemporal being in a temporal framework. This language is

considered metaphorical, but the authors of the following study

believe it may express the underlying cognitive conception of

God.

Barrett created a study in three parts, each designed to

elicit subjects’ God concepts. The first study, in part A,

evaluated differences between theological God and everyday God

concepts, by examining the differences between a theological

characterization of God as measured by a questionnaire and

concepts revealed by a story recall task. Part B looks to see if

these differences are found with other non human, super-agents.

In this condition, God was replaced with a supernatural,

futuristic computer called Uncomp. Expecting Uncomp to be

entirely novel to the participants, they could not bring any

preexisting concept to bear on the task. Study 1C was an attempt

to increase salience of theological concepts to see if

performance in story recall task could be manipulated in this

way. If so, this would be shown to be strong evidence that

24


concepts of God are what is important in determining subjects

performance and not some artifact of the task.

52 volunteer graduate students were randomly assigned to one

of each of the three studies, 22 to the first (1A) and 15 each to

Studies 1B and 1C. Materials were a questionnaire including

ratings of self religiosity and affiliation, and a number of yes-

no questions targeted at specific properties God might possess.

Subjects were asked to use their own concepts of God. Following

this, eight narratives were read on audio, 66 Recall items were

recorded, 44 that were concerned with the basic facts of the

story, and 22 God items concerned with how God was

conceptualized.

For 21 of the 44 base items, the correct answer was yes. For

the God items, the correct answer was always no, due to the

nature of the items. Relying on mistaken comprehension, all God

items included information not in the stories but suggesting

particular dimensions of an anthropomorphic concept. A yes answer

to any of these was considered evidence for the attribution of a

particular property to God that was not expressed in the story.

Study 1B was similar to 1A with the exception of the substitution

25


of Uncomp for God. In experiment 1C subjects were in the

salience condition after completing 5 free-form questions about

the relevant characteristics of God and were then given the story

recall task.

As predicted, in study 1A there was substantial inter-

subject agreement on the questionnaires, the theological concepts

describing few if any physical or psychological constraints on

God. However in the story task, there was a strong tendency for

subjects to think of God exhibiting human limitations, suggesting

an everyday anthropomorphic God concept.

Their hypothesis also held in study 1B, where subjects made

fewer errors in recalling the Uncomp items than in the previous

condition. Although the numbers were much lower, the subjects

still anthropomorphized Uncomp. After positing a variety of

explanations for the difference in the findings, the authors

suggest that anthropomorphic stories play a larger role for

stories about God than for stories about Uncomp. The computer was

not comparable to God in an important sense, God is non-natural,

and thus, Uncomp, who exists in time and space, cannot be a

proper substitute for God in this paradigm. In the 3rd study,

26


subjects as a group still exhibited poorer accuracy than

expected, based on theological beliefs as measured by the

questionnaire in Study 1 and base item accuracy. This means that

even with the priming of the theological questions, the

anthropomorphic concept was persistent.

4. Attached to a Construct

The concept of an anthropomorphic image of God leads to

another theory of the relationship between humans and their

deities. Lee Kirkpatrick offers an analogy between the attachment

theories of Bowlby (1969) and Ainsworth (1969) and the

relationships observed among the religious faithful between

themselves and God or some other sacred or supernatural figure.

Considering this, he notes that “every form of attachment

behavior, and the behavior of the attachment figure identified by

Bowlby, (1969) has it’s close counterpart in the relationship

between Israel (the worshiper) and God, ” (Reed, 1978). The

centrality of relationship themes in religious life has also been

noted by religious scholars, “just as the story of anyone’s life

is the story of relationships, so each person’s religious story

is a story of relationships.” (Greeley, 1981). In drawing the

27


line between a relationship to God and whether there is

attachment in the “clinical” sense, Kirkpatrick (2005) cites

criteria of five defining characteristics widely acknowledged to

distinguish attachment from other types of close relationships;

(1) Proximity seeking, particularly in times of stress or alarm.

Beyond infancy and early childhood, the efficacy of an

attachment figure depends not on physical, but rather

psychological proximity. Religion facilitates the feeling of

proximity to God by the foundational notion of omnipresence,

“God is always watching you”.

(2) The caregiver provides a haven of safety, Prayer and the

church have also been perceived by many to be havens of safety

during times of stress, illness or injury, (“tell your

problems to God, and he’ll comfort you”) as well as being

called upon during grieving after death.

(3) The secure base idea has a corollary in the believer’s idea

of nearness and protection by the deity illustrated by such

common sayings as “God is by my side”, “in the arms of Jesus”

“God is my rock, my fortress”. One’s relationship with the

28


deity provides a feeling of strength, self assurance, and a

sense of peace.

(4) Separation causes anxiety. Kirkpatrick (2005) cites a

variety of studies showing that the community of believers has

a lower level of anxiety due to their religious belief. In

Western, predominantly Christian samples, measures of

religious commitment correlate positively with a sense of

personal competence and control, (Ventis, 1995), an active and

flexible approach to problem solving (Pargament, Steele and

Tyler, 1979), and a sense of optimism and hope with respect to

both the long term and short term future, (Myers, 1992). At

the same time, religion has been shown to correlate inversely

with trait anxiety, (Baker & Gorsuch, 1982)

(5) The loss of the attachment figure would cause grief.

Considering this loss, Kirkpatrick (1997) looks at three

possible analogous situations. One is the loss of faith, when

people feel keenly the trauma of losing their relationship

with God. Second is when believers are forced to renounce

beliefs, such as members of a cult, who often experience

separation anxiety. The third is the theological separation

29


from God that all believers try to avoid, that of permanent

separation from God when you go to hell.

In their article “Two Dimensions of Attachment to God and

Their Relation to Affect, Religiosity, and Personality

Constructs”, Kirkpatrick and Wade Rowatt (2002) “sought to

address limitations on previous research on attachment theory and

religion by (1) developing a dimensional attachment to God scale,

and (2) demonstrating that dimensions of attachment to God are

predictive of measures of affect and personality, after

controlling for social desirability and other related dimensions

of religiosity”. Previous research done by Kirkpatrick and others

had used the measure of attachment created by Hazan &Shaver

(1987) asking respondents to select one of three responses

describing security, (God’s warmth and supportiveness), avoidance

(His distance and lack of interest), and anxious/ ambivalent (God’s

inconsistent reactions) that best characterizes his or her

beliefs about and relationship to God.

Consensus among attachment researchers is to tap two

continuous and relatively orthogonal dimensions, anxiety and

avoidance, (Brennan, Clark & Shaver, 1998, Sanford, 1997, Fraley,

30


Waller & Brennan, 2000). The authors sought to develop a self

report, multi-item, Likert-style attachment to God scale that

would parallel the current state of measurement of attachment in

other relationship domains, and examine correlates in affect and

personality.

Participants were 120 community persons (76 women, 44 men,

M age=42) and 254 undergraduates (176 women, 76 men M age=19)

from in and around the Dallas and Waco Texas areas. They were 81%

Caucasian, 7% Hispanic, 6% African-American and 6% other

ethnicity. The denominations were all Christian, the majority

Protestant, with less than 20% in either the community or

undergraduate sample reporting Catholic, Orthodox, Unitarian, or

other. Materials included 22 attachment to God items derived

from phrases from the three Hazan & Shaver paragraphs converted

to 7 point scale items (1=not at all characteristic of me; 7=very

characteristic of me.) The participants also completed 10

assessments including those of personality, affect, anxiety, and

social desirability.

Measurements of personality traits showed Neuroticism

correlated positively with both attachment to God dimensions,

31


Agreeableness and Conscientiousness correlated negatively with

each. Neither Extraversion or Openness appreciably correlated

with either attachment to God dimension.

A strong negative partial correlation was found between an

image of God as loving and the avoidance dimension of attachment

to God (r=-0.74). The avoidance dimension was also negatively

correlated with intrinsic religious orientation, doctrinal

orthodoxy, and the religious component of symbolic immortality.

The anxious attachment to God dimension was most strongly

correlated with extrinsic religious orientation (r=0.38).

Psychological results showed small positive correlations between

the anxious dimension of attachment to God, manifest anxiety, and

negative affect.

The authors found that individual difference to attachment

to God can be measured like other forms of attachment, and can be

conceptualized in terms of avoidance (vs. security) and anxiety

while acknowledging the parameters could be expanded and improved

for future research. The authors also found interesting

correlations between attachment to God dimensions and other

32


variables that support both convergent and discriminative

validity of personality, religiosity, and psychological measures.

5. Bound by a Construct Whose Intentions We Think We Know.

Perhaps we really think that Gods are people, but different,

and better. The philosopher and anthropologist Pascal Boyer

(1996) provides a description of what he refers to as the real

work of gods, that their central role is to get things done, make

things right, and keep them that way. The ontological category of

“person” is the way we think of a god, whether we assign it a

human form or not. The assumption that gods have minds is a

human, person trait exhibited by our expectation that Gods have

minds we can interact with. Human activities like prayer,

sacrifice, rituals and good behavior make sense in light of the

fact that they are done in the hope of eliciting some response

from a God. We assume they’re listening when we pray asking for

rain, notice when we’re getting married and having babies, and

and assume we will interact with them or in their domain when we

die. Boyer’s work puts forward the assertion that the special

properties of supernatural agents, the properties that make them

matter to most people, are those which directly activate mental

33


systems geared to describing and managing social interaction with

other human agents.

In their article “Beyond Beliefs: Religions Bind Individuals

into Moral Communities”

Jesse Graham and Jonathan Haidt (2010) suggest that religion

should be studied as a complex system with many social functions,

one of which is to bind people into communities. They argue that

the focus of the theories proposed above on the individual belief

in God is misguided, and that the commitment to an individualist

and cognition heavy approach may reflect a Protestant bias, since

no other religions, and not even all branches of Christianity,

show the strict division between intrinsic and extrinsic

religiosity (Cohen, Hall, Koenig, & Meador, 2005). Saying that

the true focus of religion is the creation, enacting and

maintaining of an emergent community, with God as the center of

the action, the authors quote Emile Durkheim (1915) who gave as

the definition of religion “...a unified system of belief and

practices related to sacred things...beliefs and practices which

unite into one single moral community called a church, all those

who adhere to them”. Durkheim (1915) focused on social behaviors,

34


not just individual beliefs. He also saw the importance of

collective behaviors such as rituals, practices and beliefs which

he viewed as inseparable from the “moral harmony” a member of a

congregation has and the collective actions performed for a

religious purpose.

The authors argue that the moral domain is broad, and

propose five sets of evolved moral intuitions, or foundations, on

which cultures and religions construct a great variety of virtues

and vices: (1) harm/care, (2) fairness/reciprocity, (3) in-

group/loyalty, (4) authority/respect, and (5) purity/sanctity.

The authors call the first two “individualizing foundations”

because they subserve moral systems that protect individuals and

their rights. They call the other three the “binding

foundations”, because they subserve the social functions, limit

autonomy and self expression and create emergent social entities

such as families, clans, guilds, teams, tribes, and nations.

The other half of this paper investigates three mysteries

that the authors believe can be illuminated by thinking about God

as a center of moral communities, rather than approaching

religion as a set of beliefs about God, creation and immortality.

35


It contains their predictions as to how their five Moral

Foundations relate to and influence behavior behind each mystery.

Mystery # 1: Why are religious people happier than nonreligious people?

The immense amount of data on the subject is conclusive, and

persists even when controlled for race, sex, education, income

and family status, (Brooks, 2008, et al). Social psychologists

focus on the content of religious belief as the source of

happiness, suggesting that beliefs buffer from threat and

uncertainty. The authors suggest social practices are more

important than religious beliefs, and report that the

relationship between religion and well being is driven by

religious practice, not affiliation, (Brooks, 2008). Another

study shows that lasting well being may come from integration

into a religious community, (Ysseldyk, Matheson, & Anisman,

2010). The authors predict that denominations and congregations

in which the moral order is based on the three binding

foundations will show a stronger correlation between

participation and happiness than those whose moral community

rests primarily on the harm and fairness foundations.

36


Mystery #2: Why do religious people give more to charity? Religious

people are much more likely to give money to and volunteer for

charitable causes. (Brooks, 2006, Gronjberg & Never, 2004). It

has been suggested that religious people do the contributing out

of enlightened self interest, (Harris, 2006). Others cite the

Golden Rule as the motivating force (Batson et al., 2003). For

the authors, these findings show that participation in a moral

community that values charity and selflessness increases

charitable behavior. They predict the greater sense of

connectedness in groups with a five-foundation morality make it

easier to generate giving, volunteers, and donations,

particularly when the request is made by a member of the

community. The basis of interdependence rather than autonomy of a

group leads to more willingness to part with their time and

money.

Mystery #3: Why are most people Religious? The near universality

of religious behavior has led psychologists to look for

evolutionary explanations. Whether as cognitive errors or

adaptations, belief in gods soon came under selection pressure.

In the evolutionary competitive scenario, cultures in which

37


individuals burned resources in individual worship would lose out

to atheists groups as well as atheists within their group.

Cultures that used worship to bind the group together, increase

trust, and increase the monitoring and punishment that are so

effective in suppressing cheating and free riding (Fehr &

Gachter, 2002, Gurek, Irlenbush, & Rokenbach, 2006; Haley &

Fessler, 2005) would gain a big advantage over other groups, as

well as imposing punishments on nonbelievers within the ranks

(Wilson, 2002). Such groups would meet Haidt’s (2008) definition

of a moral system; an interlocking set of values, virtues, norms,

practices, identities, institutions, technologies, and evolved

psychological systems to suppress or regulate selfishness and

make social life with a high degree of cooperation possible. In

the authors view, religious practice co-evolved with religiously

inclined minds, so that they now fit together extremely well.

III The Limitations of Current Approaches

While the evolutionary theories and explanations of

religious belief and perception as a by product of adaptation

cited in the above studies and experiments are convincing, the

theorists fail to acknowledge or explore the mystical experience

38


that is central to the lives of a certain portion of religious

people. These experiences and beliefs have helped shape the

formation of religious communities and beliefs over thousands of

years. Group or community theories, while more inclusive, simply

draws these mystics and spiritual experiencers into the crowd,

rather than suggesting that their experience is part of the

formative basis of the ideological constructs binding moral

communities.

Following the theoretical steps laid out in the above survey

of Evolutional Psychology gives us a way to think of religion in

a societal and behavioral sense, and is a valuable tool for the

field of Psychology, which has always been suspicious of the

intangibility of religion. The mystery of personal spiritual

experience, however, remains elusive.

References

Ainsworth, M.D.S. (1969). Object Relations, Dependency and Attachment: A Theoretical Review of th e Infant Mother Relationship. Child Development, Vol. 40, pp. 969-1025

Atran, S. (2002). In Gods We Trust: The Evolutionary Landscape of Religion. Oxford: Oxford University Press.

39


Baker, M., Gorsuch, R. (1982). Trait Anxiety and Intrinsic-Extrinsic religiousness. Journal for the Scientific Study of Religion. Vol. 21, pp. 119-122.

Barrett, J., Johnson, A. H. (2008). The Role of Control in Attributing Agency to Inanimate Objects. Journal of Cognition and CultureVol. 3,, pp. 208-217.

Barrett, J. L., Reichert, R., Drisenga, A. (2001). Gods Beliefs versus Mother’s: The Development of Non-human Agent Concepts. Child Development, Vol. 72, pp.50-65

Barrett, J. L., Keil, F. (1996) Conceptualizing a Non-natural Entity: Anthropomorphism in God Concepts. Cognitive Psychology Vol. 31, pp. 219-247.

Barrett, J.L., Nyhof, M. (2001). Spreading Non-natural Concepts. Journal of Cognition and Culture, Vol. 1, pp. 69-100

Batson, C. D., Lishner, D. A., Carpenter, A., Dulin, L., Harjusola-Webb, S., Stocks, E. L., et al. (2003). “. . . As you would have them do unto you”: Does imagining yourself in the other’s placestimulate moral action? Personality and Social Psychology Bulletin, 29, 1190-1201.

Bering, J. (2005). The Evolutionary History of an Illusion. In Ellis, B., Bjorklund, D. F. (Eds.) Origins of the Social Mind: Evolutionary Psychology and Child Development. Guilford Press, New York.

Bering, J., Bjorklund, D. F. (2004). The Natural Emergence of Reasoning About the Afterlife as a Developmental Regularity. Developmental Psychology, Vol. 40, 217-233

Bering, J., Baumann, B. D., (2004). Childrens Attributions of Intentions to an Invisible Agent. Manuscript Submitted for Review.

Brennan, K. A., Clark, C. L., Shaver, P. R. (1998). Self-report Measurement of Adult Attachment,: An Integrative Overview. In

40


Simpson, J. A., Rhodes, W. S. (Eds.) Attachment Theory and Close Relationships (pp. 46-76) New York: Guilford Press.

Brooks, A. C. (2008) Gross National Happiness: Why Happiness Matters for America, and How We Can Get More of It. New York: Basic Books.

Bowlby, J. (1969) Attachment and Loss: Vol. 1. Attachment. New York, BasicBooks.

Boyer, P. Religious Thought and Behaviour as By-Products of BrainFunction. Trends in Cognitive Sciences. Vol. 7, No. 3 March 2003.

Boyer, P. (1996) What makes Anthropomorphism Natural: Intuitive Ontology and Cultural Representations journal of the Royal AnthropologicalInstitute (NS) Vol. 2. pp. 83-97

Dawkins, R. (1976) The Selfish Gene. Oxford: Oxford University Press.

Durkheim, E. (1965) The Elementary Forms of Religious Life. (J.W. Swain trans.) New York: Free Press (original published in 1915).

Fehr, E., & Gachter, S. (2002). Altruistic punishment in humans. Nature, Vol. 415, 137-140.

Fiske, A. P. (1992). Four Elementary forms of Sociality: Framework for a Unified Theory of Social Relations. Psychological review Vol. 99, pp. 689-723.

Fodor, J.A. (1983). The Modularity of Mind: An Essay on Faculty Psychology. Cambridge, Mass.: MIT Press.

Fraley, R. C., Waller, N. G., Brennan, K. G. (2000). An Item response Theory Analysis of Self Report Measures of Adult Attachment. Journal of Personality and Social Psychology. Vol. 78, pp. 350-65.

41


Graham, J., Haidt, J. (2010). Beyond Beliefs: REligions Bind Individuals into Moral Communities. Personality and Social Psychology Review Vol. 14, pp. 140-152

Gould, S. J. (1991). Exaptation: A Crucial Tool for Evolutionary Psychology. Journal of Social Issues. Vol.47, No.3, pp. 43-65.

Gurek, O., Irlenbusch, B., & Rockenbach, B. (2006). The competitiveadvantage of sanctioning institutions. Science, Vol. 312, 108-111.

Guthrie, S. (1980). A Cognitive Theory of Religion. Current Anthropology, Vol. 21, no. 2, pp. 181-194

Gergely, G., Nadasdy, Z., Csibra, G. & Biro, S. (1995). Taking the intentional stance at 12 months of age. Cognition Vol. 56, 165-193.

Greeley, A. (1981). The Religious Imagination. New York: Sadlier.

Gronbjerg, K. A., & Never, B. (2004). The role of religious networks and other factors in types of volunteer work. Nonprofit Management and Leadership, Vol. 14, 263–289.

Hazan, C., Shaver, P. (1987). Romantic Love Conceptualized as an Attachment Process. Journal of Personality and Social Psychology. Vol. 52, pp. 511-524.

Haidt, J. (2008). Morality. Perspectives on Psychological Science, Vol. 3,65-72.

Haley, K. J., & Fessler, D.M.T. (2005). Nobody’s watching? Subtlecues affect generosity in an anonymous economic game. Evolution and Human Behavior, Vol. 26, 245-256.

Harris, S. (2006). Letter to a Christian nation. New York: Alfred A. Knopf.

Heider, F. & Simmel, M. (1944). An experimental study of apparent behavior. American Journal of Psychology 57, 243-249.

42


Hick, J. (2004) An Interpretation of Religion: Human Responses to the Transcendent (Second Edition). New Haven, CN: Yale University Press.

Kelemen, D. (2004). Are Children “Intuitive Theists?”: Reasoning About Purpose and Design in Nature. Psychological Science, Vol. 15, pp. 295-301

Kirkpatrick, L. (2005). Attachment, Evolution, and the Psychology of Religion., New York: Guilford Press.

Kirkpatrick, L. A. (1992). An attachment Theoretical Approach to the Psychology of Religion. International Journal for the Psychology of Religion, vol. 2, No.1, pp.3-28.

Kirkpatrick, L. A. (1997) A Longitudinal Study of Changes in Religious Belief and Behavior as a Function of Individual Differences in Adult Attachment Style. Journal for the Scientific Study of Religion. Vol. 36, pp. 207-217.

Kirkpatrick, L. A., Shaver, P. R. (1990). Attachment theory and Religion: Childhood Attachments, Religious Beliefs and Conversion. Journal for the Scientific Study of Religion, Vol. 29, pp 315-334.

Meyers, D. G. (1992). The Pursuit of Happiness. New York: Morrow.

Norenzayan, A., Atran, S. (2002). Cognitive and Emotional Processes in Cultural Transmission of Natural and Non-natural Beliefs. In Schaller, M., Crandall, C. (Eds.) The Psychological Foundations of Culture. Hillsdale , N.J.: Erlbaum.

Pargament, K.I., Steele, R. E., Tyler, F. B. (1979). Religious Participation, Religious Motivation and Individual Psychosocial Competence. Journal for the Scientific Study of Religion. Vol. 18, pp. 412-419.

Reed, B. (1978). The Dynamics of Religion: Process and Movement in Christian Churches. London: Darton, Longman and Todd.

43


Rochat, P., Morgan, R. & Carpenter, M. (1997) Young infants’ sensitivity to movement information specifying social causality. Cognitive Development 12, 537-561.

Rowatt, W. C., Kirkpatrick, L. A. (2002)Dimensions of Attachment to God and their Relation to Affect, Religiosity, and PersonalityConstructs. Journal for the Scientific Study of Religion, Vol. 41, pp. 637-651.

Sanford, K. (1997) Two Dimensions of Adult Attachment: Further Validation. Journal of Social and Personal Relationships. Vol. 14, pp. 133-43.

Shweder, R. A., Much, N. C., Mahapatra, M., & Park, L. (1997). The “big three” of morality (autonomy, community, and divinity), and the “big three” explanations of suffering. In A. Brandt & P. Rozin (Eds.), Morality and health (pp. 119-169). New York: Routledge.

Ventis, W.L. (1995). The Relationships between Religion and Mental Health. Journal of social Issues, vol. 51, no. 2, pp.33-48.

Wilson, D. S. (2002). Darwin’s cathedral: Evolution, religion, and the Nature of Society. Chicago: University of Chicago Press.

Wilson, D. S., Van Vogt, M., O’Gorman, R. Testing Major Evolutionary Hypotheses about Religion with a Random Sample. Human Nature, Winter, 2005 Vol. 16, No. 4, pp.382-409

44

Religion and Evolutionary Psychology

Documents