There's a particular, hard-to-name ache in realizing that a friendship only activates under certain conditions — usually theirs, usually bad ones. You're glad to hear from them. You also notice, quietly, that you never call first and be met with the same warmth you extend when things are calm.
Before deciding what that means about the friendship, it's worth being honest about what it doesn't mean: it doesn't mean you're imagining it, and it also doesn't mean your memory of "how often" is a reliable instrument.
Closeness has to be maintained, and maintenance isn't distributed evenly
Robin Dunbar's research on social group structure describes friendship not as a fixed state but as a set of concentric layers — a small inner circle, a wider circle, an outer one — each requiring a certain frequency of contact to stay at that level of closeness.1 Later work extended this to show the layers are real and measurable, not just a metaphor.2 A friendship that only gets contact during a crisis is, structurally, receiving one kind of maintenance and none of the other kind — and layers that don't get maintained drift outward, whether or not anyone intends it.
You can't trust your own memory of "how often"
This is the part that's easy to skip past. Hermann Ebbinghaus's foundational work on forgetting showed that memory for specific instances decays fastest in the period right after they happen3 — which means your felt sense of "we talk all the time" or "they never reach out" is being assembled, after the fact, from a badly incomplete sample, weighted toward whatever's recent. Recency does a lot of quiet editing on a story that feels like a settled fact.
Roles get divided, and division isn't always mutual
Daniel Wegner's work on transactive memory describes how close relationships divide up cognitive labor — who remembers the plans, who initiates contact, who tracks the emotional temperature.4 Some of that division is fair. Some of it just calcifies because one person happened to initiate first a few times and it became the default. Mihaly Csikszentmihalyi and Reed Larson's experience-sampling studies found that people's in-the-moment reports of how they're spending time with others diverge substantially from their later, summarized account of the same relationship5 — another reason a dated, contemporaneous record beats a retrospective feeling when the question is genuinely "who initiates, and how often."
None of this tells you what to do about the friend who only calls when things are bad. It does tell you that the fair way to settle the question isn't to relitigate your memory of it — it's to look at the actual pattern, dated, and let that answer instead. If this is on your mind, you might also read the people you stopped mentioning or why some friendships end without a fight.
- Dunbar, R. I. M. (1993). Coevolution of neocortical size, group size and language in humans. Behavioral and Brain Sciences, 16(4), 681–694.
- Dunbar, R. I. M. (1998). The social brain hypothesis. Evolutionary Anthropology, 6(5), 178–190.
- Ebbinghaus, H. (1885). Über das Gedächtnis: Untersuchungen zur experimentellen Psychologie. Leipzig: Duncker & Humblot.
- Wegner, D. M. (1987). Transactive memory: A contemporary analysis of the group mind. In B. Mullen & G. R. Goethals (Eds.), Theories of Group Behavior (pp. 185–208). New York: Springer. See also Wegner, D. M. (1995). A computer network model of human transactive memory. Social Cognition, 13(3), 319–339.
- Csikszentmihalyi, M., & Larson, R. (1987). Validity and reliability of the experience-sampling method. Journal of Nervous and Mental Disease, 175(9), 526–536.