In sta­tis­tics, one learns very early on that inde­pen­dent ran­dom vari­a­tions tend to can­cel out when grouped together. By the law of large num­bers, the col­lec­tive behav­ior of many inde­pen­dent ran­dom enti­ties tends to closely reflect the aver­age behav­ior of these enti­ties. This is very con­ve­nient if we see the ran­dom vari­a­tions as being mean­ing­less noise and were only inter­ested in the aver­age. But if the vari­a­tions con­tained all the inter­est­ing com­plex­ity, then the com­plex­ity is washed away in the aggre­gate. In a generic soci­ety, the gener­ics may be highly var­ied and indi­vid­u­ally very com­plex. Naively put them in a group and the col­lec­tive behav­ior is sim­pler – quite plau­si­bly, the vari­a­tions can­cel toward zero and the group achieves noth­ing as a whole.

This all changes if the ran­dom vari­a­tions were not inde­pen­dent and tended to align along cer­tain dimen­sions. The vari­a­tions will be ampli­fied wher­ever they align, and the col­lec­tive behav­ior of the group cleanly empha­sizes the align­ment of its con­stituents. If we want a group of gener­ics to retain a mean­ing­ful iden­tity dis­tinct from the aver­age of its mem­bers, we need to give the gener­ics a desire to align with each other in behav­ior or moti­va­tion. The topic of today’s post is the coherency domain, which con­tains ideas that lend well to being mixed with other con­structs to describe non­triv­ial social behav­iors. As an exam­ple, I will use con­cepts from the coherency domain to describe the gam­ing com­bos, which are a set of meth­ods for cre­at­ing align­ment even between unre­lated or mutu­ally exclu­sive activ­i­ties through the com­mon par­tic­i­pa­tion of a big­ger event.

The Coherency Domain

The cen­tral con­cept in the coherency domain is that of per­ceived eval­u­a­tion sim­i­lar­ity – the belief that one eval­u­ates a scene in a sim­i­lar way as other gener­ics (even if their roles in the scene are dis­sim­i­lar). Two gener­ics may have dif­fer­ent senses / eval­u­a­tions / agents, but if there are high-level com­mon­al­i­ties in the vis­i­ble con­se­quences of their judg­ments then they may res­onate accord­ing to the met­ric of per­ceived eval­u­a­tion sim­i­lar­ity. For a more tech­ni­cal descrip­tion, let’s intro­duce the senses Coherency-Resonant and Coherency-Antiresonant. The for­mer cap­tures the idea of per­ceived eval­u­a­tion sim­i­lar­ity, and the lat­ter cap­tures the feel­ing of being left out from a res­o­nant group. Coherency-resonant can be defined like:

define Coherency-Resonant(scene): (scene -> boolean) as
  let generics = GetGenerics(scene),
  let my_causal_estimate, my_evaluation =
    EventInterpretation(SplitPriorPosterior(scene)),
  if there is some generic in generics where
    Mirror(scene, generic) = Success(mirrored_scene)
  then
    let their_causal_estimate, their_evaluation =
      EventInterpretation(SplitPriorPosterior(mirrored_scene)),
    if IsSimilar(my_evaluation, their_evaluation) then true
    otherwise false
  otherwise false

The def­i­n­i­tion here has some nested lay­ers but it mostly uses con­cepts we already dis­cussed in pre­vi­ous posts. First, we find the other gener­ics that appear in a scene, and we split the scene into its prior and pos­te­rior infor­ma­tion so that we can use event inter­pre­ta­tion to get an eval­u­a­tion for the scene. Then we try to mir­ror the scene to the other gener­ics, and when suc­cess­ful we use event inter­pre­ta­tion on the mir­rored scenes to get the eval­u­a­tions from the other gener­ics’ per­spec­tives. If there is sig­nif­i­cant sim­i­lar­ity between the eval­u­a­tion from the ori­gin generic’s per­spec­tive and the eval­u­a­tion from a dif­fer­ent per­spec­tive, then the sense of Coherency-Resonant is triggered.

I believe there is a need to empha­size sev­eral points about this def­i­n­i­tion. Firstly, this whole thought process hap­pens inside one generic’s mind, and gen­er­ally it doesn’t know the actual thought processes of other gener­ics. The ori­gin generic can only make edu­cated guesses based on his knowl­edge. Secondly, when per­form­ing event inter­pre­ta­tion on the mir­rored scene the ori­gin generic doesn’t nec­es­sar­ily assume that all other gener­ics think the same way he does. The encod­ing of the mir­rored scene is still aware that the scene is being viewed from a dif­fer­ent individual’s per­spec­tive, and if the ori­gin generic had a recur­sive model of this other indi­vid­ual then his lens can use this recur­sive model to pro­duce eval­u­a­tions based on the other individual’s ten­den­cies. Lastly, the com­par­i­son here is being made between eval­u­a­tions instead of senses. Evaluations can be more com­plex and may depend on spe­cific details in the scene or even cer­tain mem­o­ries in the expe­ri­ence foun­da­tion¹. Though the def­i­n­i­tion of Coherency-Resonant is rel­a­tively sim­ple in the grand scheme of things, the con­di­tions for trig­ger­ing this sense can be quite nuanced due to the com­plex­ity hid­den in the recur­sive mod­els and in the eval­u­a­tions being pro­duced by the lens.

The other sense in the coherency domain, though named Coherency-Antiresonant, doesn’t sim­ply mean the lack of Coherency-Resonant. In fact the two senses are not mutu­ally exclu­sive. Coherency-Antiresonant describes a per­ceived mis­align­ment in eval­u­a­tions that were suc­cess­fully aligned in other indi­vid­u­als. It can be defined like:

define Coherency-Antiresonant(scene): (scene -> boolean) as
  let generics = GetGenerics(scene),
  let my_causal_estimate, my_evaluation =
    EventInterpretation(SplitPriorPosterior(scene)),
  if there is some generic in generics where
    Mirror(scene, generic) = Success(mirrored_scene), and
    Coherency-Resonant(mirrored_scene)
  then
    let their_causal_estimate, their_evaluation =
      EventInterpretation(SplitPriorPosterior(mirrored_scene)),
    if not IsSimilar(my_evaluation, their_evaluation) then true
    otherwise false
  otherwise false

Like the Coherency-Resonant sense, in this def­i­n­i­tion we find other gener­ics in the scene and use event inter­pre­ta­tion + the Mirror func­tion to deter­mine the eval­u­a­tions from mul­ti­ple per­spec­tives. However, we require that the eval­u­a­tions be dis­sim­i­lar (note the not before the IsSimilar), and also that the Coherency-Resonant sense be suc­cess­fully trig­gered from some other generic’s per­spec­tive. This usu­ally implies the exis­tence of at least three par­tic­i­pants in the scene – the ori­gin generic, the generic who is sub­ject to com­par­i­son in this def­i­n­i­tion, and some third generic who res­onates with the sec­ond². It’s pos­si­ble to have four gener­ics in the scene and cre­ate a two vs. two sce­nario, where each pair of gener­ics are res­o­nant with each other but not with the other pair. Under this scene, the gener­ics would expe­ri­ence both the Coherency-Resonant and the Coherency-Antiresonant senses.

If most indi­vid­u­als in an envi­ron­ment have an agent favor­ing the first-per­son Coherency-Resonant eval­u­a­tion, then we may see gener­ics group­ing together in ways that empha­size nat­u­rally-occur­ring align­ments in thought processes. Merely being in the pres­ence of a res­o­nant group can alter one’s lens to more closely resem­ble those of the group, since the observer will wit­ness many behav­iors that are well-explained by the shared eval­u­a­tions, and there­fore his lens will pro­duce sim­i­lar eval­u­a­tions more often to improve its pre­dic­tive accu­racy. Since infor­ma­tion about the motives of gener­ics make up a sig­nif­i­cant por­tion of a lens’s eval­u­a­tion, we’d expect to find res­o­nance in any group of gener­ics engag­ing in sim­i­lar activ­i­ties with sim­i­lar motives.

Those who are seek­ing the Coherency-Antiresonant sense may have coun­ter­cul­ture ten­den­cies, and may even form groups of indi­vid­u­als specif­i­cally try­ing to be dif­fer­ent from other groups. On the other hand, gener­ics may have neg­a­tive feed­back loop agents that are ener­gized by the first-per­son Coherency-Antiresonant eval­u­a­tion but work toward reduc­ing the pro­duc­tion of the Coherency-Antiresonant sense. The wishes of such an agent can be achieved triv­ially by avoid­ing those who pre­fer dis­sim­i­lar activ­i­ties / have dis­sim­i­lar thought processes. But the agent can also achieve its goal in a non­triv­ial way by mak­ing other gener­ics recon­tex­tu­al­ize their activ­i­ties to empha­size an eval­u­a­tion shared by all par­tic­i­pants. It is pos­si­ble that cre­ative think­ing on the part of the Coherency-Antiresonant min­i­mizer agent is what led to the dis­cov­ery of the some­what intri­cate yet very com­mon­place game com­bos, which are use­ful for cre­at­ing res­o­nance even when indi­vid­u­als engage in dif­fer­ent activ­i­ties or act with dif­fer­ent motivations.

The Cooperative Game Combo

One method of intro­duc­ing res­o­nance between gener­ics with­out chang­ing their behav­iors too much is by invent­ing a com­mon goal that encom­passes mul­ti­ple exist­ing behav­iors. This com­mon goal, also known as the coop­er­a­tive game, could sim­ply be the cel­e­bra­tion of a fes­ti­val which requires per­form­ing a set of actions as rit­u­als, or it can be the com­ple­tion of a com­pos­ite task that can only be done through the actions of mul­ti­ple indi­vid­u­als of diverse roles³. In either form, the coop­er­a­tive game is designed so that exist­ing behav­iors in a group of gener­ics con­ve­niently accom­plish var­i­ous sub-goals of the over­ar­ch­ing com­mon goal. By doing this, highly diverse behav­iors are recon­tex­tu­al­ized as actions that serve a com­mon goal, and all indi­vid­u­als par­tic­i­pat­ing in the game may feel the Coherency-Resonant sense due to a per­ceived sim­i­lar­ity in motives. When activ­i­ties with dis­sim­i­lar motives are able to gen­er­ate res­o­nance due to the exis­tence of a coop­er­a­tive game, we say that the var­i­ous motives and the desire to achieve res­o­nance have come together to form a coop­er­a­tive game combo.

In order for the coop­er­a­tive game combo to exist, we gen­er­ally need the par­tic­i­pants of the game to sat­isfy a num­ber of conditions:

1. The gener­ics are moti­vated to gen­er­ate the Coherency-Resonant sense. 
   • Without this con­di­tion, the gener­ics would be per­fectly con­tent doing their activ­i­ties in iso­la­tion with­out need­ing a fes­ti­val / com­pos­ite task.
2. The gener­ics are moti­vated to engage in activ­i­ties that are some­what inde­pen­dent, i.e. the activ­i­ties do not mutu­ally exclude each other, but work­ing on one activ­ity does not require or imply work­ing on the other activities. 
   • Clearly it’s not pos­si­ble to com­plete the com­mon goal if some activ­i­ties are mutu­ally exclu­sive. But if activ­i­ties tend to require or encap­su­late each other, then there is no need to intro­duce an exter­nal struc­ture to tie the activ­i­ties together.
3. The gener­ics are capa­ble of suc­cess­fully per­form­ing their activ­i­ties with­out too much dif­fi­culty or risk of failure. 
   • The coop­er­a­tive game is only suc­cess­ful if all its con­stituent activ­i­ties are suc­cess­ful. One fail­ure can ruin the expe­ri­ence for oth­ers, so the par­tic­i­pants are moti­vated to exclude exces­sively dif­fi­cult activ­i­ties from the combo.

As an inter­est­ing con­se­quence, par­tic­i­pants in a coop­er­a­tive game combo could be moti­vated to assist with (or even directly par­tic­i­pate in) other activ­i­ties of the coop­er­a­tive game, even if they orig­i­nally didn’t feel any urge to per­form such other activ­i­ties. The suc­cess of all activ­i­ties is rel­e­vant due to their con­tri­bu­tion toward the over­ar­ch­ing com­mon goal, so par­tic­i­pants would still expe­ri­ence res­o­nance when assist­ing or dab­bling in these other activ­i­ties. This ten­dency helps spread aware­ness and inter­est in the activ­i­ties of the coop­er­a­tive game, which would cre­ate future oppor­tu­ni­ties for res­o­nance sim­ply by encour­ag­ing larger groups of gener­ics to par­take in those activ­i­ties. The coop­er­a­tive game basi­cally acts like an effi­cient cat­a­lyst for the rapid pro­duc­tion of the Coherency-Resonant sense.

The Competitive Game Combo

Let’s sup­pose we had a group of gener­ics whose behav­iors fol­low the dynam­ics of the Rivalry domain. For instance, they may be act­ing to max­i­mize the pro­duc­tion of the Rivalry-Selfish sense by try­ing to cre­ate the sense of Gain while simul­ta­ne­ously cre­at­ing the sense of Pain in oth­ers. Usually, two com­pet­ing gener­ics can’t both feel the Rivalry-Selfish sense since both play­ers require their rival to “lose”. By the very nature of their motives, it is com­mon for their activ­i­ties to be mutu­ally exclu­sive, and there­fore their activ­i­ties can’t both be included in a coop­er­a­tive game combo⁴. However, there is another method of cre­at­ing res­o­nance even between com­pet­ing play­ers of a Rivalry-heavy envi­ron­ment. One can cre­ate a com­mon task called a com­pet­i­tive game which sim­ply asks for its par­tic­i­pants to try their best at achiev­ing their own mutu­ally exclu­sive goals. Of course, most par­tic­i­pants of the com­pet­i­tive game will fail and only the cham­pion of the game suc­ceeds. However, all par­tic­i­pants have suc­ceeded in at least attempt­ing to achieve their goal, and they may feel the sense of Coherency-Resonant since they all appear to have a sim­i­lar moti­va­tion of try­ing their best to become the champion.

In this way, even though the par­tic­i­pants of the game engage in essen­tially the same activ­i­ties with or with­out the com­pet­i­tive game struc­ture, the exis­tence of the game has recon­tex­tu­al­ized the activ­i­ties so that the goal is sim­ply a best-effort par­tic­i­pa­tion in the game. Regardless of suc­cess or fail­ure, engag­ing in the activ­i­ties pro­duces an eval­u­a­tion related to this new goal of sim­ply par­tic­i­pat­ing in the game, and all gener­ics will be able to res­onate on the com­mon­al­ity of this eval­u­a­tion. When activ­i­ties with mutu­ally exclu­sive goals (usu­ally Rivalry-related) are able to come together and pro­duce res­o­nance through the exis­tence of a com­pet­i­tive game, we say that the motives of the mutu­ally exclu­sive goals and the desire to achieve res­o­nance have come together to form a com­pet­i­tive game combo.

Like the coop­er­a­tive game combo, there are a num­ber of con­di­tions that we usu­ally need for the com­pet­i­tive game combo to work correctly:

1. Participants in the game believe that they are all try­ing to achieve mutu­ally exclu­sive goals. 
   • If the par­tic­i­pants do not believe this, then they might not believe that they are all tak­ing the com­pet­i­tive game seri­ously or that they all needed to spend sig­nif­i­cant effort to par­tic­i­pate in the game. This ruins the sense of res­o­nance between the players.
2. Participants are moti­vated to gen­er­ate the Coherency-Resonant sense. 
   • If par­tic­i­pants did not care about res­o­nance then they do not need a com­pet­i­tive game. They would be fine with sim­ply attempt­ing to achieve their goals in isolation.
3. The par­tic­i­pants’ moti­va­tion to find res­o­nance is greater than their moti­va­tion to avoid failure. 
   • A com­pet­i­tive game tends to attract strong rivals, so gener­ics who are strongly moti­vated to suc­ceed in their goals would avoid par­tic­i­pat­ing in such games even if they desired res­o­nance⁵.

It’s pos­si­ble to cre­ate an extra layer of res­o­nance by group­ing play­ers in a com­pet­i­tive game into teams. If the over­all goals of the teams are mutu­ally exclu­sive, then the same prin­ci­ples apply and all par­tic­i­pants in the com­pet­i­tive game will be res­o­nant in their shared moti­va­tion of try­ing their best to make their team win. However, there is also res­o­nance between team mem­bers after each suc­cess or fail­ure in the game, as each per­son per­ceives the other team mem­bers as hav­ing the same eval­u­a­tion of the sit­u­a­tion. It is pos­si­ble to take this idea one step fur­ther and design the com­pet­i­tive game so that each team must com­plete a com­pos­ite task (thereby forc­ing the team mem­bers to engage in diverse activ­i­ties). By doing this, the mem­bers of each team would be par­tic­i­pat­ing in a coop­er­a­tive game nested inside a com­pet­i­tive game. This cre­ates res­o­nance in at least three dif­fer­ent ways as the gener­ics par­tic­i­pate in the over­all team vs. team com­pet­i­tive game.

Closing Words

Cooperative games and com­pet­i­tive games both have fairly intri­cate rules / designs, so it seems strange that they would be com­mon­place in a generic soci­ety. But we only need one indi­vid­ual to invent these designs in order for the other mem­bers of a Coherency-dom­i­nated generic soci­ety to rec­og­nize the util­ity of the game com­bos. Those observ­ing the activ­i­ties of the games (or related arti­facts such as prepa­ra­tion or spec­ta­tor­ship of the games⁶) may expe­ri­ence the Coherency-Antiresonant sense, and those who dis­like the antires­o­nance may choose to take part one way or another. In this way, the games tend to adver­tise them­selves and aware­ness of the games rapidly spreads across the generic soci­ety. Perhaps this explains how even ancient human civ­i­liza­tions were able to come up with coop­er­a­tive & com­pet­i­tive game designs, and how the game com­bos ended up being so com­mon­place in human society.

1. I won­der if this post is feel­ing like a big mash-up of all the pre­vi­ous posts, con­sid­er­ing all the older tech­ni­cal terms being brought up here. I guess it was a good idea to bold and ital­i­cize the terms after all.
2. Under a com­pli­cated setup where the recur­sive model of the other generic itself has an inac­cu­rate recur­sive model of the ori­gin generic, it might be pos­si­ble to trig­ger Coherency-Antiresonant with just two par­tic­i­pants. I think this vari­a­tion would still behave sim­i­larly to the nor­mal ver­sion with three or more par­tic­i­pants, but let me know if you have any thoughts about this.
3. Examples of fes­ti­vals in human soci­ety include events like school tal­ent shows, hol­i­day cel­e­bra­tions and char­ity fairs. Examples of com­pos­ite tasks include pro­duc­ing plays / movies, or play­ing music in a band / orchestra.
4. Here we assume the def­i­n­i­tion of the activ­ity requires suc­cess­fully trig­ger­ing Rivalry-Selfish. If so, it’s prob­lem­atic to include even the activ­i­ties of one player in a coop­er­a­tive game, since it’s dif­fi­cult to guar­an­tee suc­cess if the rival has sim­i­lar or higher skill. On the other hand, if we apply the com­pet­i­tive game combo and recon­tex­tu­al­ize the activ­ity so that only par­tic­i­pa­tion is required (i.e. win­ning is not nec­es­sary), then there is no prob­lem includ­ing the activ­i­ties of one or more com­pet­ing play­ers into a greater coop­er­a­tive game.
5. Usually in com­pet­i­tive games found in human soci­ety, the goal of the game is to assert some level of con­trol over the oppo­nent in a restricted envi­ron­ment. For most games (i.e. any­thing but the world cham­pi­onships) the inher­ent reward of win­ning is small and the inher­ent pun­ish­ment of los­ing is small, so it seems like the com­pet­i­tive game is designed to max­i­mize the num­ber of will­ing par­tic­i­pants and to cre­ate the feel­ing of res­o­nance in a pure form. However, some peo­ple react strongly to the feel­ing of los­ing con­trol, and would be moti­vated not to play the com­pet­i­tive game even if there was no other con­se­quence for defeat.
6. Such arti­facts can them­selves be con­sid­ered part of a greater fes­ti­val that con­tains other games within it.