TL;DR
Singing together is one of the fastest-known ways to bond a group of strangers, and the effect is measurable down to real endorphin release, not just a feeling of closeness.12 Repeating devotional lyrics for months or years then does two more things at once: the act of singing gives the mind a behavior it has to explain to itself, and the words themselves get more convincing purely from being heard again and again.34 None of this requires the singer to have believed anything on day one — belief is a plausible side effect of the practice, not a precondition for it.
Ayyappa devotees in traditional black pilgrimage attire on the path to Sabarimala — a tradition built around group bhajan singing throughout the trek. Photo by Kandukuru Nagarjun, CC BY 2.0.5
The Puzzle of the Reluctant Convert
It’s a common shape: someone joins a devotional singing group for the company, the music, or a family obligation, holding no real conviction about the words. Years later, they’re genuinely devout. The intuitive explanations reach for either mysticism or manipulation. Neither is necessary. There’s a well-studied, entirely mundane set of mechanisms that explains the shift, and none of them care what deity is named in the lyrics.
The Body Bonds Before the Mind Agrees
Start with what singing does to a room of strangers, measured in the body rather than the mind. Robin Dunbar and colleagues had members of the Oxford University Boat Club row on ergometers, once alone and once in a synchronized group of six, and measured pain tolerance afterward as a proxy for endorphin release. Rowing in sync roughly doubled the increase in pain threshold compared to rowing solo — 11.04 mmHg versus 5.63 mmHg — a real, physiological signature of the “high” synchronized effort produces, not just a nicer mood.1
Singing appears to be an unusually efficient trigger for that same synchrony effect. A separate study followed new adult-education classes — some singing groups, some doing crafts or creative writing — over seven months, tracking self-reported closeness at the start, middle, and end. The singing groups bonded dramatically faster: a large jump in felt closeness after the very first session, before the craft and writing groups had built any comparable connection at all.2 The researchers called it an “ice-breaker effect” — singing collapses the usual months-long timeline for stranger-to-community bonding into something closer to a single session.
Acting Becomes a Reason to Believe
That bonded feeling needs somewhere to go, and this is where the mind’s own bookkeeping takes over. Psychologist Daryl Bem’s self-perception theory describes a pattern that runs backward from how people assume attitudes work: when someone’s internal feelings about something are ambiguous, they infer their own attitude by watching their own behavior, the same way they’d read a stranger’s attitude from that stranger’s actions.3 Someone who has been showing up weekly, then monthly, then almost daily, to sing words of surrender and devotion has generated a large body of “devotional behavior” for their own mind to explain. The simplest, most consistent explanation the mind can reach for is: I must actually believe this. The behavior doesn’t just express a belief. Repeated for long enough, it starts to manufacture one.
Repetition Makes the Words Feel True
The lyrics themselves get a separate boost purely from being heard again and again. Robert Zajonc’s mere-exposure research showed that repeated contact with a completely neutral, meaningless stimulus — invented words, unfamiliar characters — measurably increases how positively people rate it, with no other information added.4 There’s a real limit to this, though, and it’s a genuine complication for the “sing it enough and you’ll believe it forever” version of the story: a 2017 meta-analysis found the relationship isn’t a straight line upward but an inverted U, where liking builds for the first ten to twenty exposures and then plateaus or even reverses with genuine overexposure.4 Repetition is a real lever on conviction, but it’s not an infinite one.
It Isn’t Reserved for the Devout
None of these three mechanisms is specific to religious practice. A separate line of research on synchronized movement and cooperation makes this explicit, noting that “armies, churches, organizations, and communities” all rely on the same trick — marching, chanting, and praying in unison — to push group members into synchrony and, through that synchrony, into deeper cooperation.6 Marching soldiers, chanting sports fans, and praying congregations are running the identical synchrony-to-cooperation pathway, just pointed at different objects of loyalty. A stadium chant and a temple bhajan are, mechanically, close cousins: both compress strangers into a bonded group fast, both hand the brain a repeated behavior to rationalize into a belief, and both make their own words feel truer the more times they’re sung.
What the Mechanism Doesn’t Decide
None of this says whether the belief that grows this way is true, healthy, or worth keeping — the mechanism is neutral about content. It explains how conviction can grow reliably out of shared repeated practice with no initial belief required, which is a different question from whether any particular conviction, once grown, deserves to be kept. That second question still has to be answered by looking at what the belief actually asks of the person holding it — the singing only explains how they got there, not whether it was worth the trip.
