Rubicon report

Is loneliness a vicious circle?

Question Do these interviewees describe loneliness as a vicious circle, and do their accounts show support breaking into it?

Six interviews drawn at random (seed 7) from the London youth loneliness interviews under 40,000 characters. Rubric, set before any interview was read: a vicious circle is present if at least 3 of the 6 interviewees each tell a reinforcing loop through Loneliness whole, in their own account; support breaks in if at least 2 interviewees tell a support or service link into such a loop with a weakening sign, or tell a balancing loop through Loneliness whole. The answer makes no claim about which loop dominates.

Vicious circle: not shown, and we are fairly sure of that. None of the six interviewees tells a reinforcing loop through loneliness whole in their own account (, , , ), where the evaluator's standard needs at least three, so the vicious circle is not shown (). Reinforcing loops through loneliness appear only when links from different interviewees are joined. The nearest thing to a circle in one person's words is 's account of depression leading to withdrawal and time alone leading back to depression (), but it runs through being alone rather than through feeling lonely, and he tells it of other people ("I wouldn’t know, I’ve never been depressed").

6 interviews in 2 groups · 70 passages coded · 13 October 2026

  • Verification in this report:
  • 35 numbers, every one computed directly from the coded passages rather than written by the AI. Click any number to see who it counts.
  • 70 quotations, every one found word for word in its interview. Click an interview's ID beside a quotation to read it in place.
  • 47 findings and 4 high-level findings, every one checked against the interviews after they were coded a second time.

Support breaking in: shown by the standard, but it says less than it seems. interviewees (, , , and ) tell a link from support into one of the joined-up reinforcing loops with a weakening sign, where the standard needs two, so support breaking in is shown. The count clears the bar by enough that reading a link or two the other way would not change it, so the verdict itself is fairly secure. What it shows is weaker: the support is almost all from family and friends, most of these accounts describe support that was missing or not sought making things worse rather than support arriving and easing them, and nobody tells whole the loops it reaches. tells no link from support at all. The other route in the standard, a balancing loop told whole, is met by one interviewee only ().

Loneliness is told as an effect, seldom as a cause

All the interviewees but say plainly what brings loneliness on, each at least partly of their own life, and hints at it. Less time with other people is the commonest cause (, and 1 more only hints at it): "you need to put yourself around people to not feel lonely" (), or, of the City, "It just feels lonely ‘cause there’s no one I know there" (). says the same of why he is not lonely himself: "The reason why I don’t feel lonely is because there’s so many different ways of communication nowadays" (). Others name missing support from family and friends (, and 1 more only hints at it) (, , ), being misunderstood or not sharing others' experience (): "I’ve always had people around me but I’ve been more misunderstood, which has led to being lonely" (), and, in 's account, not knowing how to talk to people: "having people to communicate with but not knowing how to do it leads to kind of loneliness" ().

Loneliness as a cause of something else is rarer. , from his own trips alone, says that "being lonely is what causes being scared or anxious or nervous in those situations" (), and , of people in general, links depression to the feeling "you just feel like, you feel alone" () (). says that after a breakup, "because the person doesn’t want to experience the feeling of loneliness, they might engage in it", meaning a new relationship () (). A loop through loneliness needs a link out of it as well as one into it, and none of the interviewees who tell a link out also tell a path back that closes a reinforcing loop.

The map below shows every signed link the interviewees tell, with the number of interviewees behind each; links that sit in no loop are kept.

Drawing the map with Graphviz, which loads from the internet the first time a report is opened.

Each arrow is a causal link; its number is how many sources mention it. The sign at each arrowhead says whether the two move the same way (+), opposite ways (−) or the source did not say (?). Click an arrow or a box to read its passages.

holds the documents’ full text

The reinforcing loops are joined up from separate accounts

Four reinforcing loops through loneliness appear in the combined links, and nobody tells any of them whole. The diagrams below draw each loop the answer names, with the links into it from outside.

Drawing the map with Graphviz, which loads from the internet the first time a report is opened.

B1: a balancing loop, told whole in interviewees. Each arrow is numbered by the interviewees telling that link; a dashed box is a variable outside the loop that drives it. Click an arrow or a factor for its passages.

holds the documents’ full text

Drawing the map with Graphviz, which loads from the internet the first time a report is opened.

B2: a balancing loop, told whole in interviewees. Each arrow is numbered by the interviewees telling that link; a dashed box is a variable outside the loop that drives it. Click an arrow or a factor for its passages.

holds the documents’ full text

Drawing the map with Graphviz, which loads from the internet the first time a report is opened.

R1: a reinforcing loop, told whole in interviewees. Each arrow is numbered by the interviewees telling that link; a dashed box is a variable outside the loop that drives it. Click an arrow or a factor for its passages.

holds the documents’ full text

Drawing the map with Graphviz, which loads from the internet the first time a report is opened.

R2: a reinforcing loop, told whole in interviewees. Each arrow is numbered by the interviewees telling that link; a dashed box is a variable outside the loop that drives it. Click an arrow or a factor for its passages.

holds the documents’ full text

Drawing the map with Graphviz, which loads from the internet the first time a report is opened.

R3: a reinforcing loop, told whole in interviewees. Each arrow is numbered by the interviewees telling that link; a dashed box is a variable outside the loop that drives it. Click an arrow or a factor for its passages.

holds the documents’ full text

Drawing the map with Graphviz, which loads from the internet the first time a report is opened.

R4: a reinforcing loop, told whole in interviewees. Each arrow is numbered by the interviewees telling that link; a dashed box is a variable outside the loop that drives it. Click an arrow or a factor for its passages.

holds the documents’ full text

  • R1, loneliness lowering mental wellbeing and low mood raising loneliness, joins 's and 's links from loneliness to fear or depression () (, ) to 's link from depression to loneliness () (). Interviewees who tell it whole: .
  • R2 adds withdrawal. comes closest to telling it: "when you’re depressed, you just don’t wanna talk to no one" (), and of pushing people away, "You make yourself lonely- lonelier every day." (). That is a spiral in his words, but it runs from depression through withdrawal to loneliness; he does not say that loneliness deepens the depression, so the loop closes only with another interviewee's link. The link from low mood to withdrawal is his alone (), and interviewees who tell the loop whole: .
  • R3 adds self-worth, through a link from contact to self-worth that nobody states firmly (, and 2 hint at it): of a reserved classmate left feeling "maybe he is a weirdo" (), and of a breakup leaving people feeling "that person has left your life, that you’re worthless" (). Interviewees who tell it whole: .
  • R4 adds help from those around you, through 's link that being around people means "you get help" () (). Interviewees who tell it whole: .

These loops are a reasonable draft of how loneliness might sustain itself among these young people, but they are the analyst's assembly of separate accounts, and they need testing with people who live it before they are treated as a finding.

Support breaks in, mostly as its absence

The standard asks whether support or services weaken such a loop. § The reinforcing loops are joined up from separate accounts In these accounts the support is almost all from family and friends, and mostly it is told as something missing. Its absence brings loneliness for interviewees, and 1 more hints at it: says "I just felt like I was alone being at my mum’s." (), says people "probably feel most alone when you need the support of others but can’t get it" (), and describes his own choice not to ask: "instead of me opening up and seeking help, I’m like ‘forget it, I’ll do it alone’" (). Missing support also lowers self-worth for interviewees, and 1 more hints at it: 's friends not inviting her out "just made me feel really unwanted" (), and , putting in effort her family did not return, says "feel like and idiot" ().

Support that is present and easing things appears in 's account, where her sisters make the family rifts bearable: "it doesn’t really bother me as much, ‘cause I’ve still got my sisters" (). She is the only interviewee to tell support easing low mood (); the other who hints at that link, , says only that having no one to turn to can "feel quite sad" (1) ().

Services barely appear. says that if he were depressed and "no one offered me the help, no one offered me the numbers or stuff like that to call, the help lines, I would feel hurt" () (), and he and only hint that services ease loneliness itself (2): that "everyone that is depressed should talk to someone. To not be lonely in the future." (), and that the bodies responsible for rough sleepers "don’t necessary care that much. And that can bring feelings of loneliness" (), which concerns services failing other people rather than him. mentions his own counsellor, but says that with his friends "I still feel alone", so the counselling is not told as changing his loneliness.

tells a balancing loop through loneliness whole, B1 (): loneliness after a breakup leads people into a new relationship, and people overlook a new partner's faults "because they stop them from feeling lonely" (). He tells it of other people, and adds that the new partner "might not be appropriate", so it is a loop that eases loneliness rather than an account of support working well. No interviewee tells a second balancing loop, B2, whole in explicit links (); 's account would close it only through his hint that a breakup leaves people feeling worthless (1) (), so B1 is the only balancing loop anyone tells whole.

Limitations

The six interviews were drawn at random from the shorter interviews in the study, and holds only the second part of its interview. The interviews asked what people associate with loneliness and where they feel most connected and most lonely, not how it changes over time, so they invite accounts of causes rather than of cycles; a vicious circle that interviewees live but are not asked to trace would not show in these links. Every loop rests on the analyst's reading of which of eleven variables a passage concerns, and the reinforcing loops exist only by joining different people's accounts. Nothing here says which loop dominates or how strong any of them is.

Annexes

How this report was written

What was read. Rubicon read all 6 documents in full. The counts below are of interviewees, one document each.

Coding. It then restated each causal claim the documents make as a link from a cause to an effect, each end one of a list of 11 factors, and whether the effect goes up or down.

Second coding and check. The same documents were coded again to the same definitions; the run does not record whether a separate coder did it, so it is not called independent here. The two codings differed in 22 places. A reader ruled on each against the text: 19 for the first coding, 3 for the second. It tightened 2 definitions as a result, and corrected the coding to match. It searched the documents each count leaves out for passages both coders had missed, and added them. It also checked every sentence and quotation of the answer against the documents, so the coding and the answer in this report are the corrected ones, counted again after the check.

Counting. Interviewees were then counted in 2 tables: a causal map of every link; the feedback loops of up to 4 links through Loneliness.

Verdicts. Rubicon reached its verdicts by rules agreed beforehand. They come from the evaluator's rubric, set before any interview was read. For each question the rule asks, the answer was counted from the tables and set against the thresholds the rule gives: “Do at least 3 of the 6 interviewees each tell, in their own account, a reinforcing loop through Loneliness whole?” reached Not shown; “Do at least 2 interviewees tell a support or service link into a reinforcing loop through Loneliness with a weakening sign, or tell a balancing loop through Loneliness whole?” reached Shown.

Writing and checks. Rubicon wrote the answer from these counts and verdicts, citing for each finding the passages it rests on. Every number in the answer, 35 in all, was computed directly from the coded passages rather than written by the AI, and all 70 quotations were found word for word in their documents.

Revisions. The report was revised twice after it was first drawn, and counted and checked again each time; the changes are listed at the end of this annex.

The workflow x-ray beside Step by step, below, draws the same procedure, a box for each stage and what it made.

Step by step

The analysis was written down as a workflow before it was counted: what to look for, how to code it, what to count. It was then counted directly from the coded passages rather than by the AI, so anyone can check a number, and another coder can follow the same definitions and code the interviews afresh.

What was read

All 6 documents, read in full: 4 male, 2 female.

Counting: Map

Interviewees by cause and effect and sign

The table 28 cells
CauseEffectSignInterviewees
Social contactLonelinessMinus of 6 +1 implicit
Going out and taking partSocial contactPlus of 6
Support from family and friendsLonelinessMinus of 6 +1 implicit
BelongingLonelinessMinus of 6
LonelinessMental wellbeingMinus of 6
Social contactMental wellbeingPlus of 6 +1 implicit
Support from family and friendsSelf-worthPlus of 6 +1 implicit
BelongingMental wellbeingPlus of 6
BelongingSelf-worthPlus of 6
Going out and taking partBelongingPlus of 6 +1 implicit
Going out and taking partLonelinessMinus of 6
Going out and taking partMental wellbeingPlus of 6
LonelinessSocial contactPlus of 6
Mental wellbeingLonelinessMinus of 6
Mental wellbeingSocial contactPlus of 6
Self-worthLonelinessMinus of 6 +1 implicit
Social confidenceLonelinessMinus of 6 +1 implicit
Social confidenceSocial contactPlus of 6
Social contactSupport from family and friendsPlus of 6
Support from family and friendsBelongingPlus of 6
Support from family and friendsMental wellbeingPlus of 6 +1 implicit
Support from family and friendsSocial contactPlus of 6
Support from servicesMental wellbeingPlus of 6
Work or studyLonelinessMinus of 6
Social confidenceMental wellbeingPlus of 6 +1 implicit
Social contactSelf-worthPlus of 6 +2 implicit
Support from servicesLonelinessMinus of 6 +2 implicit
Work or studySocial contactMinus of 6 +1 implicit

holds the documents’ full text

Counting: Loops

Interviewees by cause and effect

A path, a chain somebody told or a loop counts as explicit for a document only where explicit links alone can make it.

The table 196 cells
LoopPolarityPartLinkInterviewees
Loneliness →+ Social contact →- LonelinessBWhole of 6
Loneliness →+ Social contact →- LonelinessBLinkLoneliness →+ Social contact of 6
Loneliness →+ Social contact →- LonelinessBLinkSocial contact →- Loneliness of 6 +1 implicit
Loneliness →+ Social contact →- LonelinessBDriverBelonging →- Loneliness of 6
Loneliness →+ Social contact →- LonelinessBDriverGoing out and taking part →- Loneliness of 6
Loneliness →+ Social contact →- LonelinessBDriverGoing out and taking part →+ Social contact of 6
Loneliness →+ Social contact →- LonelinessBDriverMental wellbeing →- Loneliness of 6
Loneliness →+ Social contact →- LonelinessBDriverMental wellbeing →+ Social contact of 6
Loneliness →+ Social contact →- LonelinessBDriverSelf-worth →- Loneliness of 6 +1 implicit
Loneliness →+ Social contact →- LonelinessBDriverSocial confidence →- Loneliness of 6 +1 implicit
Loneliness →+ Social contact →- LonelinessBDriverSocial confidence →+ Social contact of 6
Loneliness →+ Social contact →- LonelinessBDriverSupport from family and friends →- Loneliness of 6 +1 implicit
Loneliness →+ Social contact →- LonelinessBDriverSupport from family and friends →+ Social contact of 6
Loneliness →+ Social contact →- LonelinessBDriverSupport from services →- Loneliness of 6 +2 implicit
Loneliness →+ Social contact →- LonelinessBDriverWork or study →- Loneliness of 6
Loneliness →+ Social contact →- LonelinessBDriverWork or study →- Social contact of 6 +1 implicit
Loneliness →+ Social contact →+ Self-worth →- LonelinessBWhole of 6 +1 implicit
Loneliness →+ Social contact →+ Self-worth →- LonelinessBLinkLoneliness →+ Social contact of 6
Loneliness →+ Social contact →+ Self-worth →- LonelinessBLinkSocial contact →+ Self-worth of 6 +2 implicit
Loneliness →+ Social contact →+ Self-worth →- LonelinessBLinkSelf-worth →- Loneliness of 6 +1 implicit
Loneliness →+ Social contact →+ Self-worth →- LonelinessBDriverBelonging →- Loneliness of 6
Loneliness →+ Social contact →+ Self-worth →- LonelinessBDriverBelonging →+ Self-worth of 6
Loneliness →+ Social contact →+ Self-worth →- LonelinessBDriverGoing out and taking part →- Loneliness of 6
Loneliness →+ Social contact →+ Self-worth →- LonelinessBDriverGoing out and taking part →+ Social contact of 6
Loneliness →+ Social contact →+ Self-worth →- LonelinessBDriverMental wellbeing →- Loneliness of 6
Loneliness →+ Social contact →+ Self-worth →- LonelinessBDriverMental wellbeing →+ Social contact of 6
Loneliness →+ Social contact →+ Self-worth →- LonelinessBDriverSocial confidence →- Loneliness of 6 +1 implicit
Loneliness →+ Social contact →+ Self-worth →- LonelinessBDriverSocial confidence →+ Social contact of 6
Loneliness →+ Social contact →+ Self-worth →- LonelinessBDriverSupport from family and friends →- Loneliness of 6 +1 implicit
Loneliness →+ Social contact →+ Self-worth →- LonelinessBDriverSupport from family and friends →+ Self-worth of 6 +1 implicit
Loneliness →+ Social contact →+ Self-worth →- LonelinessBDriverSupport from family and friends →+ Social contact of 6
Loneliness →+ Social contact →+ Self-worth →- LonelinessBDriverSupport from services →- Loneliness of 6 +2 implicit
Loneliness →+ Social contact →+ Self-worth →- LonelinessBDriverWork or study →- Loneliness of 6
Loneliness →+ Social contact →+ Self-worth →- LonelinessBDriverWork or study →- Social contact of 6 +1 implicit
Belonging →- Loneliness →+ Social contact →+ Support from family and friends →+ BelongingBWhole of 6
Belonging →- Loneliness →+ Social contact →+ Support from family and friends →+ BelongingBLinkBelonging →- Loneliness of 6
Belonging →- Loneliness →+ Social contact →+ Support from family and friends →+ BelongingBLinkLoneliness →+ Social contact of 6
Belonging →- Loneliness →+ Social contact →+ Support from family and friends →+ BelongingBLinkSocial contact →+ Support from family and friends of 6
Belonging →- Loneliness →+ Social contact →+ Support from family and friends →+ BelongingBLinkSupport from family and friends →+ Belonging of 6
Belonging →- Loneliness →+ Social contact →+ Support from family and friends →+ BelongingBDriverGoing out and taking part →+ Belonging of 6 +1 implicit
Belonging →- Loneliness →+ Social contact →+ Support from family and friends →+ BelongingBDriverGoing out and taking part →- Loneliness of 6
Belonging →- Loneliness →+ Social contact →+ Support from family and friends →+ BelongingBDriverGoing out and taking part →+ Social contact of 6
Belonging →- Loneliness →+ Social contact →+ Support from family and friends →+ BelongingBDriverMental wellbeing →- Loneliness of 6
Belonging →- Loneliness →+ Social contact →+ Support from family and friends →+ BelongingBDriverMental wellbeing →+ Social contact of 6
Belonging →- Loneliness →+ Social contact →+ Support from family and friends →+ BelongingBDriverSelf-worth →- Loneliness of 6 +1 implicit
Belonging →- Loneliness →+ Social contact →+ Support from family and friends →+ BelongingBDriverSocial confidence →- Loneliness of 6 +1 implicit
Belonging →- Loneliness →+ Social contact →+ Support from family and friends →+ BelongingBDriverSocial confidence →+ Social contact of 6
Belonging →- Loneliness →+ Social contact →+ Support from family and friends →+ BelongingBDriverSupport from services →- Loneliness of 6 +2 implicit
Belonging →- Loneliness →+ Social contact →+ Support from family and friends →+ BelongingBDriverWork or study →- Loneliness of 6
Belonging →- Loneliness →+ Social contact →+ Support from family and friends →+ BelongingBDriverWork or study →- Social contact of 6 +1 implicit
Loneliness →- Mental wellbeing →- LonelinessRWhole of 6
Loneliness →- Mental wellbeing →- LonelinessRLinkLoneliness →- Mental wellbeing of 6
Loneliness →- Mental wellbeing →- LonelinessRLinkMental wellbeing →- Loneliness of 6
Loneliness →- Mental wellbeing →- LonelinessRDriverBelonging →- Loneliness of 6
Loneliness →- Mental wellbeing →- LonelinessRDriverBelonging →+ Mental wellbeing of 6
Loneliness →- Mental wellbeing →- LonelinessRDriverGoing out and taking part →- Loneliness of 6
Loneliness →- Mental wellbeing →- LonelinessRDriverGoing out and taking part →+ Mental wellbeing of 6
Loneliness →- Mental wellbeing →- LonelinessRDriverSelf-worth →- Loneliness of 6 +1 implicit
Loneliness →- Mental wellbeing →- LonelinessRDriverSocial confidence →- Loneliness of 6 +1 implicit
Loneliness →- Mental wellbeing →- LonelinessRDriverSocial confidence →+ Mental wellbeing of 6 +1 implicit
Loneliness →- Mental wellbeing →- LonelinessRDriverSocial contact →- Loneliness of 6 +1 implicit
Loneliness →- Mental wellbeing →- LonelinessRDriverSocial contact →+ Mental wellbeing of 6 +1 implicit
Loneliness →- Mental wellbeing →- LonelinessRDriverSupport from family and friends →- Loneliness of 6 +1 implicit
Loneliness →- Mental wellbeing →- LonelinessRDriverSupport from family and friends →+ Mental wellbeing of 6 +1 implicit
Loneliness →- Mental wellbeing →- LonelinessRDriverSupport from services →- Loneliness of 6 +2 implicit
Loneliness →- Mental wellbeing →- LonelinessRDriverSupport from services →+ Mental wellbeing of 6
Loneliness →- Mental wellbeing →- LonelinessRDriverWork or study →- Loneliness of 6
Loneliness →- Mental wellbeing →+ Social contact →- LonelinessRWhole of 6
Loneliness →- Mental wellbeing →+ Social contact →- LonelinessRLinkLoneliness →- Mental wellbeing of 6
Loneliness →- Mental wellbeing →+ Social contact →- LonelinessRLinkMental wellbeing →+ Social contact of 6
Loneliness →- Mental wellbeing →+ Social contact →- LonelinessRLinkSocial contact →- Loneliness of 6 +1 implicit
Loneliness →- Mental wellbeing →+ Social contact →- LonelinessRDriverBelonging →- Loneliness of 6
Loneliness →- Mental wellbeing →+ Social contact →- LonelinessRDriverBelonging →+ Mental wellbeing of 6
Loneliness →- Mental wellbeing →+ Social contact →- LonelinessRDriverGoing out and taking part →- Loneliness of 6
Loneliness →- Mental wellbeing →+ Social contact →- LonelinessRDriverGoing out and taking part →+ Mental wellbeing of 6
Loneliness →- Mental wellbeing →+ Social contact →- LonelinessRDriverGoing out and taking part →+ Social contact of 6
Loneliness →- Mental wellbeing →+ Social contact →- LonelinessRDriverSelf-worth →- Loneliness of 6 +1 implicit
Loneliness →- Mental wellbeing →+ Social contact →- LonelinessRDriverSocial confidence →- Loneliness of 6 +1 implicit
Loneliness →- Mental wellbeing →+ Social contact →- LonelinessRDriverSocial confidence →+ Mental wellbeing of 6 +1 implicit
Loneliness →- Mental wellbeing →+ Social contact →- LonelinessRDriverSocial confidence →+ Social contact of 6
Loneliness →- Mental wellbeing →+ Social contact →- LonelinessRDriverSupport from family and friends →- Loneliness of 6 +1 implicit
Loneliness →- Mental wellbeing →+ Social contact →- LonelinessRDriverSupport from family and friends →+ Mental wellbeing of 6 +1 implicit
Loneliness →- Mental wellbeing →+ Social contact →- LonelinessRDriverSupport from family and friends →+ Social contact of 6
Loneliness →- Mental wellbeing →+ Social contact →- LonelinessRDriverSupport from services →- Loneliness of 6 +2 implicit
Loneliness →- Mental wellbeing →+ Social contact →- LonelinessRDriverSupport from services →+ Mental wellbeing of 6
Loneliness →- Mental wellbeing →+ Social contact →- LonelinessRDriverWork or study →- Loneliness of 6
Loneliness →- Mental wellbeing →+ Social contact →- LonelinessRDriverWork or study →- Social contact of 6 +1 implicit
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRWhole of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRLinkLoneliness →- Mental wellbeing of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRLinkMental wellbeing →+ Social contact of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRLinkSocial contact →+ Self-worth of 6 +2 implicit
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRLinkSelf-worth →- Loneliness of 6 +1 implicit
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRDriverBelonging →- Loneliness of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRDriverBelonging →+ Mental wellbeing of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRDriverBelonging →+ Self-worth of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRDriverGoing out and taking part →- Loneliness of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRDriverGoing out and taking part →+ Mental wellbeing of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRDriverGoing out and taking part →+ Social contact of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRDriverSocial confidence →- Loneliness of 6 +1 implicit
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRDriverSocial confidence →+ Mental wellbeing of 6 +1 implicit
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRDriverSocial confidence →+ Social contact of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRDriverSupport from family and friends →- Loneliness of 6 +1 implicit
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRDriverSupport from family and friends →+ Mental wellbeing of 6 +1 implicit
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRDriverSupport from family and friends →+ Self-worth of 6 +1 implicit
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRDriverSupport from family and friends →+ Social contact of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRDriverSupport from services →- Loneliness of 6 +2 implicit
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRDriverSupport from services →+ Mental wellbeing of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRDriverWork or study →- Loneliness of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Self-worth →- LonelinessRDriverWork or study →- Social contact of 6 +1 implicit
Loneliness →- Mental wellbeing →+ Social contact →+ Support from family and friends →- LonelinessRWhole of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Support from family and friends →- LonelinessRLinkLoneliness →- Mental wellbeing of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Support from family and friends →- LonelinessRLinkMental wellbeing →+ Social contact of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Support from family and friends →- LonelinessRLinkSocial contact →+ Support from family and friends of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Support from family and friends →- LonelinessRLinkSupport from family and friends →- Loneliness of 6 +1 implicit
Loneliness →- Mental wellbeing →+ Social contact →+ Support from family and friends →- LonelinessRDriverBelonging →- Loneliness of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Support from family and friends →- LonelinessRDriverBelonging →+ Mental wellbeing of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Support from family and friends →- LonelinessRDriverGoing out and taking part →- Loneliness of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Support from family and friends →- LonelinessRDriverGoing out and taking part →+ Mental wellbeing of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Support from family and friends →- LonelinessRDriverGoing out and taking part →+ Social contact of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Support from family and friends →- LonelinessRDriverSelf-worth →- Loneliness of 6 +1 implicit
Loneliness →- Mental wellbeing →+ Social contact →+ Support from family and friends →- LonelinessRDriverSocial confidence →- Loneliness of 6 +1 implicit
Loneliness →- Mental wellbeing →+ Social contact →+ Support from family and friends →- LonelinessRDriverSocial confidence →+ Mental wellbeing of 6 +1 implicit
Loneliness →- Mental wellbeing →+ Social contact →+ Support from family and friends →- LonelinessRDriverSocial confidence →+ Social contact of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Support from family and friends →- LonelinessRDriverSupport from services →- Loneliness of 6 +2 implicit
Loneliness →- Mental wellbeing →+ Social contact →+ Support from family and friends →- LonelinessRDriverSupport from services →+ Mental wellbeing of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Support from family and friends →- LonelinessRDriverWork or study →- Loneliness of 6
Loneliness →- Mental wellbeing →+ Social contact →+ Support from family and friends →- LonelinessRDriverWork or study →- Social contact of 6 +1 implicit
Loneliness →+ Social contact →+ Mental wellbeing →- LonelinessBWhole of 6
Loneliness →+ Social contact →+ Mental wellbeing →- LonelinessBLinkLoneliness →+ Social contact of 6
Loneliness →+ Social contact →+ Mental wellbeing →- LonelinessBLinkSocial contact →+ Mental wellbeing of 6 +1 implicit
Loneliness →+ Social contact →+ Mental wellbeing →- LonelinessBLinkMental wellbeing →- Loneliness of 6
Loneliness →+ Social contact →+ Mental wellbeing →- LonelinessBDriverBelonging →- Loneliness of 6
Loneliness →+ Social contact →+ Mental wellbeing →- LonelinessBDriverBelonging →+ Mental wellbeing of 6
Loneliness →+ Social contact →+ Mental wellbeing →- LonelinessBDriverGoing out and taking part →- Loneliness of 6
Loneliness →+ Social contact →+ Mental wellbeing →- LonelinessBDriverGoing out and taking part →+ Mental wellbeing of 6
Loneliness →+ Social contact →+ Mental wellbeing →- LonelinessBDriverGoing out and taking part →+ Social contact of 6
Loneliness →+ Social contact →+ Mental wellbeing →- LonelinessBDriverSelf-worth →- Loneliness of 6 +1 implicit
Loneliness →+ Social contact →+ Mental wellbeing →- LonelinessBDriverSocial confidence →- Loneliness of 6 +1 implicit
Loneliness →+ Social contact →+ Mental wellbeing →- LonelinessBDriverSocial confidence →+ Mental wellbeing of 6 +1 implicit
Loneliness →+ Social contact →+ Mental wellbeing →- LonelinessBDriverSocial confidence →+ Social contact of 6
Loneliness →+ Social contact →+ Mental wellbeing →- LonelinessBDriverSupport from family and friends →- Loneliness of 6 +1 implicit
Loneliness →+ Social contact →+ Mental wellbeing →- LonelinessBDriverSupport from family and friends →+ Mental wellbeing of 6 +1 implicit
Loneliness →+ Social contact →+ Mental wellbeing →- LonelinessBDriverSupport from family and friends →+ Social contact of 6
Loneliness →+ Social contact →+ Mental wellbeing →- LonelinessBDriverSupport from services →- Loneliness of 6 +2 implicit
Loneliness →+ Social contact →+ Mental wellbeing →- LonelinessBDriverSupport from services →+ Mental wellbeing of 6
Loneliness →+ Social contact →+ Mental wellbeing →- LonelinessBDriverWork or study →- Loneliness of 6
Loneliness →+ Social contact →+ Mental wellbeing →- LonelinessBDriverWork or study →- Social contact of 6 +1 implicit
Loneliness →+ Social contact →+ Support from family and friends →- LonelinessBWhole of 6
Loneliness →+ Social contact →+ Support from family and friends →- LonelinessBLinkLoneliness →+ Social contact of 6
Loneliness →+ Social contact →+ Support from family and friends →- LonelinessBLinkSocial contact →+ Support from family and friends of 6
Loneliness →+ Social contact →+ Support from family and friends →- LonelinessBLinkSupport from family and friends →- Loneliness of 6 +1 implicit
Loneliness →+ Social contact →+ Support from family and friends →- LonelinessBDriverBelonging →- Loneliness of 6
Loneliness →+ Social contact →+ Support from family and friends →- LonelinessBDriverGoing out and taking part →- Loneliness of 6
Loneliness →+ Social contact →+ Support from family and friends →- LonelinessBDriverGoing out and taking part →+ Social contact of 6
Loneliness →+ Social contact →+ Support from family and friends →- LonelinessBDriverMental wellbeing →- Loneliness of 6
Loneliness →+ Social contact →+ Support from family and friends →- LonelinessBDriverMental wellbeing →+ Social contact of 6
Loneliness →+ Social contact →+ Support from family and friends →- LonelinessBDriverSelf-worth →- Loneliness of 6 +1 implicit
Loneliness →+ Social contact →+ Support from family and friends →- LonelinessBDriverSocial confidence →- Loneliness of 6 +1 implicit
Loneliness →+ Social contact →+ Support from family and friends →- LonelinessBDriverSocial confidence →+ Social contact of 6
Loneliness →+ Social contact →+ Support from family and friends →- LonelinessBDriverSupport from services →- Loneliness of 6 +2 implicit
Loneliness →+ Social contact →+ Support from family and friends →- LonelinessBDriverWork or study →- Loneliness of 6
Loneliness →+ Social contact →+ Support from family and friends →- LonelinessBDriverWork or study →- Social contact of 6 +1 implicit
Loneliness →+ Social contact →+ Support from family and friends →+ Mental wellbeing →- LonelinessBWhole of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Mental wellbeing →- LonelinessBLinkLoneliness →+ Social contact of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Mental wellbeing →- LonelinessBLinkSocial contact →+ Support from family and friends of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Mental wellbeing →- LonelinessBLinkSupport from family and friends →+ Mental wellbeing of 6 +1 implicit
Loneliness →+ Social contact →+ Support from family and friends →+ Mental wellbeing →- LonelinessBLinkMental wellbeing →- Loneliness of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Mental wellbeing →- LonelinessBDriverBelonging →- Loneliness of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Mental wellbeing →- LonelinessBDriverBelonging →+ Mental wellbeing of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Mental wellbeing →- LonelinessBDriverGoing out and taking part →- Loneliness of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Mental wellbeing →- LonelinessBDriverGoing out and taking part →+ Mental wellbeing of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Mental wellbeing →- LonelinessBDriverGoing out and taking part →+ Social contact of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Mental wellbeing →- LonelinessBDriverSelf-worth →- Loneliness of 6 +1 implicit
Loneliness →+ Social contact →+ Support from family and friends →+ Mental wellbeing →- LonelinessBDriverSocial confidence →- Loneliness of 6 +1 implicit
Loneliness →+ Social contact →+ Support from family and friends →+ Mental wellbeing →- LonelinessBDriverSocial confidence →+ Mental wellbeing of 6 +1 implicit
Loneliness →+ Social contact →+ Support from family and friends →+ Mental wellbeing →- LonelinessBDriverSocial confidence →+ Social contact of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Mental wellbeing →- LonelinessBDriverSupport from services →- Loneliness of 6 +2 implicit
Loneliness →+ Social contact →+ Support from family and friends →+ Mental wellbeing →- LonelinessBDriverSupport from services →+ Mental wellbeing of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Mental wellbeing →- LonelinessBDriverWork or study →- Loneliness of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Mental wellbeing →- LonelinessBDriverWork or study →- Social contact of 6 +1 implicit
Loneliness →+ Social contact →+ Support from family and friends →+ Self-worth →- LonelinessBWhole of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Self-worth →- LonelinessBLinkLoneliness →+ Social contact of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Self-worth →- LonelinessBLinkSocial contact →+ Support from family and friends of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Self-worth →- LonelinessBLinkSupport from family and friends →+ Self-worth of 6 +1 implicit
Loneliness →+ Social contact →+ Support from family and friends →+ Self-worth →- LonelinessBLinkSelf-worth →- Loneliness of 6 +1 implicit
Loneliness →+ Social contact →+ Support from family and friends →+ Self-worth →- LonelinessBDriverBelonging →- Loneliness of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Self-worth →- LonelinessBDriverBelonging →+ Self-worth of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Self-worth →- LonelinessBDriverGoing out and taking part →- Loneliness of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Self-worth →- LonelinessBDriverGoing out and taking part →+ Social contact of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Self-worth →- LonelinessBDriverMental wellbeing →- Loneliness of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Self-worth →- LonelinessBDriverMental wellbeing →+ Social contact of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Self-worth →- LonelinessBDriverSocial confidence →- Loneliness of 6 +1 implicit
Loneliness →+ Social contact →+ Support from family and friends →+ Self-worth →- LonelinessBDriverSocial confidence →+ Social contact of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Self-worth →- LonelinessBDriverSupport from services →- Loneliness of 6 +2 implicit
Loneliness →+ Social contact →+ Support from family and friends →+ Self-worth →- LonelinessBDriverWork or study →- Loneliness of 6
Loneliness →+ Social contact →+ Support from family and friends →+ Self-worth →- LonelinessBDriverWork or study →- Social contact of 6 +1 implicit

holds the documents’ full text

Judging: Verdict

The evaluator's rubric, set before any interview was read.

  • Do at least 3 of the 6 interviewees each tell, in their own account, a reinforcing loop through Loneliness whole? · level met: Not shown
  • Do at least 2 interviewees tell a support or service link into a reinforcing loop through Loneliness with a weakening sign, or tell a balancing loop through Loneliness whole? · level met: Shown
The verdicts the rule could reach
ShownThe accounts meet the standard for this part of the question.
Not shownThe accounts do not meet the standard for this part of the question.

Writing: Answer

The instruction the answer was written to

The checks that were run

  • 70 of 70 quotations found word for word in their document.
  • 35 numbers in the answer, every one computed directly from the coded passages rather than written by the AI; 0 written by the AI without a count behind them.
  • 0 citations to passages that do not exist; 0 quotations not in the passage they cite.

Second coding

The same documents were coded again to the same definitions; the run does not record whether a separate coder did it, so it is not called independent here.

The second coder's rows
  • Links: 49 rows

In the run's zip, under check/coded/.

The check

A reader ruled on every place the two codings differed, and checked every sentence and quotation of the answer against the documents, correcting the coding and the answer where they were wrong. The answer above is the corrected one, recounted after the check.

The check's record check.md
Counts
  • Disagreements between the two coders, out of 22 listed in recode/disagreements.md: ruled for the rows as coded, 19; ruled for the second coder, 3 (one of them, RVIEW2 cause Loneliness, only in part); left unsettled, 0. Two of the 22 were settled by tightening a definition (Belonging against Social contact) rather than by the existing wording.
  • Rows changed, out of the 61 rows checked: 5 rows corrected (3 causes changed, 2 weak marks removed) and 9 rows added, so coded/links.json now holds 70 rows. No row was deleted.
  • Definitions changed in workflow.json: 2 values (Social contact and Belonging, in both the cause and the effect lists). Rubric thresholds and levels unchanged.
  • Sentences of the report changed, out of the 36 in the original: 31 changed or cut (several only because cell ids renumbered when the rows changed), 5 kept word for word; the corrected report has 42 sentences, 37 of them new or rewritten.
  • Verdicts: vicious circle Not shown (0 of 6, unchanged); support breaking in Shown (5 of 6, unchanged). The wording of both now says how sure, and the support verdict now says what it does and does not show.
Definition tightened
  • Social contact and Belonging, IPANT4, VIEW1M, RVIEW2. The old definitions did not say where being among strangers goes, or where being misunderstood goes. Now: being among strangers with no one they know there to be with or talk to is Social contact; being understood by, and sharing experience with, the people around them (or feeling disconnected from a group or community they count as theirs) is Belonging. Every document was re-read for both kinds of passage; the recoding is the IPANT4 and RVIEW2/VIEW1M changes below. MBER22's "teenagers understand each other" passage was read and not coded, since it links understanding to feeling connected, not to any of the eleven variables' links into loneliness, mood or self-worth.
Rulings on the disagreements
  • IPANT4, cause Belonging (first coder) against Social contact (second): "you don't know anyone when you’re on the tube", "it's basically literally being alone although there’s people around you, you don't know them", "It just feels lonely ‘cause there’s no one I know there". For the second coder. He is describing being among strangers with nobody he knows to talk to (the tube, Liverpool Street), not a community he belongs to; under the tightened definition this is Social contact. Rows jvsdq, 26wy7, 6s32f recoded.
  • IPANT4, cause Social confidence: "could have make you scared of maybe interacting with someone that you don't know ... maybe that can make you lonely I guess". For the rows, kept weak. Fear of approaching strangers is Social confidence; it stays weak because the "that" is loose and, asked again a few lines later, he says the fear follows the loneliness ("I’d say that way around").
  • MBER21, cause Support from services: "everyone that is depressed should talk to someone. To not be lonely in the future." and "no one offered me the help ... the help lines, I would feel hurt". For the rows. Asked whether he means friends or "external support", he answers "But professionals. Definitely", and help lines are services. The first stays weak (advice, and "someone" can be a friend); the second is a stated link to hurt, which is low mood.
  • RVIEW2, cause Belonging: "eventually, you’d become lonely because you feel like no one understands you". For the rows, and made firm: the passage states the link, and under the tightened definition being understood is Belonging.
  • RVIEW2, cause Self-worth: "it would make someone feel like they’re irrelevant ... could really lead to being lonely". For the rows, kept weak: feeling irrelevant is self-worth, but he puts it among an "accumulation" of feelings.
  • RVIEW2, cause Support from family and friends: "instead of me opening up and seeking help, I’m like ‘forget it, I’ll do it alone’". For the rows: he says not asking his friends for help builds the "bubble" of loneliness, which is less help received leading to more loneliness.
  • RVIEW2, cause Work or study: "with doing that you kind of pull away from other people ... yourself in your room working towards your goal". For the rows, kept weak: "doing that" is pursuing his business single-mindedly; the second coder's reading of the same passage as contact leading to loneliness is already a row (82d42).
  • RVIEW2, cause Loneliness (second coder only): "so loneliness, sadness, depression everything comes from that" and "that could definitely lead to loneliness as well, and that could leave the person feeling low". In part for the second coder. In the first passage "that" is the self-sufficient "bubble", so loneliness is an effect there, not a cause; coded instead as Social contact to Mental wellbeing, weak, since the bubble is the cause of the sadness and depression (row 8uxjw). In the second, "that could leave the person feeling low" can refer to the loneliness just named, so Loneliness to Mental wellbeing was added, weak (row 6psmh).
  • VIEW1M, cause Belonging: "misunderstood, which has led to being lonely", "lack of shared experience can lead to the feeling of loneliness", "wonders why they can’t fit in ... feeling of inadequacy", "in my local community ... I don’t feel like I know those people". For the rows. All four are about fitting in or being understood, which the tightened definition makes Belonging; the first, his own experience stated as a cause, is now firm.
  • VIEW1M, cause Going out and taking part: "if I didn’t go to the barbershop, I wouldn’t have half as many people to…think football specifically". For the rows: going to the barber's gives him people to talk to.
  • VIEW1M, cause Loneliness: "because the person doesn’t want to experience the feeling of loneliness, they might engage in it". For the rows: read with the sentence before ("some people a year after their break up they’re still feeling pretty lonely. So anyone can come in"), loneliness after a breakup leads people into a new relationship, which is more contact.
  • YPART2, cause Social contact: "I think in the past it's made me feel quite lonely". For the rows, kept weak: the interviewer asked about friends posting nights out she was not invited to, and "it" points back to that, but loosely.
  • MBER22, effect Mental wellbeing (first) against Loneliness (second): "it doesn’t really bother me as much, ‘cause I’ve still got my sisters". For the rows: the sentence before is "before, it did make me actually feel sad", so what her sisters ease is sadness.
  • MBER22, effect Self-worth: "you put all the effort in and then they don’t do nothing in return. So, feel like and idiot". For the rows: feeling like an idiot, and earlier "a mug", is how she values herself.
  • MBER22, effect Social contact: "it doesn’t seem like they care about me so what’s the point in me putting in the effort". For the rows: less care from her family leads her to make less effort to see them.
  • RVIEW2, effect Self-worth: "so they don’t really approach them, so that can lead the individual feeling like maybe he is a weirdo". For the rows, kept weak: the cause is partly others' labelling, not only being left alone.
  • VIEW1M, effect Self-worth: "can’t fit in ... feeling of inadequacy" and "not really having people to lean on ... ask questions on yourself". For the rows; the second stays weak, since asking questions of yourself only hints at lower self-worth.
  • VIEW1M, effect Social contact: the breakup and barbershop passages. For the rows, as above.
  • YPART2, effect Loneliness: "made me feel quite lonely". For the rows, weak, as above.
  • MBER21, effect Support from family and friends (second coder only): "putting yourself around people is just the best way to live because you get help". For the second coder: being around family, friends and colleagues brings help. Row 24fca added.
  • VIEW1M, sign plus on four rows (fit in to inadequacy; no one to lean on to questioning yourself; loneliness to a new relationship; barbershop to people). For the rows: in each the two variables move the same way.
  • YPART2, sign minus on "made me feel quite lonely". For the rows: less contact, more loneliness.
Omissions found
  • Loneliness to Mental wellbeing ("Loneliness as a cause ... one interviewee says"): RVIEW2 also tells it, "probably even depression, ‘cause ... you just feel like, you feel alone" (row rn5at, firm) and "that could leave the person feeling low" (row 6psmh, weak). The link out of loneliness is now told by IPANT4 and RVIEW2, so R1's first link no longer rests on one interviewee, and the report's opening claim that two links of R2 "rest on one interviewee each" was wrong.
  • Support from family and friends to Loneliness ({map.c3}, MBER22, RVIEW2, VIEW1M): MBER21 hints at it, "instead of getting some support- moral support from someone else. So that’s how I associate loneliness" (row a9u54, weak).
  • Social contact to Self-worth (the report: "one interviewee only hints at"): VIEW1M hints at it too, "you don’t feel like that person has left your life, that you’re worthless" (row y55ya, weak). This also makes a second balancing loop, B2, whole in VIEW1M's account, through that weak link; the report now says so.
  • Support from family and friends to Mental wellbeing (the report: sisters, one interviewee): VIEW1M, asked how it feels when no one comes to help, "Um, it can feel quite sad. Um, bitter maybe." (row a9d6s, weak). MBER22 remains the only firm account.
  • Belonging to Loneliness: VIEW1M, "if they don’t quite understand your lived experience, you can feel quite alone" (row hpfjs), and Belonging to Mental wellbeing: RVIEW2, "It could lead to you feeling low, definitely because it’s like regardless of what I say, people are not gonna understand me" (row wathw). Both follow from the tightened Belonging definition; the first changes no count.
  • Social contact to Support from family and friends: MBER21 (row 24fca, from the second coder). It closes a new joined-up reinforcing loop, R4, which nobody tells whole.
  • Searched and not found: a link from low mood to withdrawal (R2) outside MBER21; a link from loneliness back into withdrawal in any account other than VIEW1M's breakup passage; help from services easing loneliness in any account other than MBER21's and VIEW1M's hints. YPART2's turning to dating apps to make up for "how little those interactions are" was read and not coded, because both ends are time with other people.
Changes to the report
  • Opening: rewritten to state both verdicts first, each with how sure. The vicious-circle verdict cited only three of the four reinforcing loops' "whole" cells; it now cites all four.
  • Opening: "two of them rest on one interviewee each" cut; with RVIEW2's link added, only the link from low mood to withdrawal rests on one interviewee (MBER21).
  • Opening: support verdict now names the five interviewees behind it and IPANT4 as the one who tells no support link, says the count is well clear of the standard's two, and says that most of these accounts are of support missing or not sought, so the verdict says little about support arriving and breaking a circle. "Support does break into these linked-up loops" and "easing loneliness" were overstated as written.
  • Opening: added MBER21's own spiral (depression, withdrawal, being alone, depression), told whole but through time alone rather than through feeling lonely, and told of others ("I wouldn’t know, I’ve never been depressed").
  • "Almost every interviewee explains what brings loneliness on": corrected to all but YPART2, who only hints at it, and says each tells it at least partly of their own life.
  • "not fitting in or not knowing the people around them ({map.c4}, and ... more only weakly) [links.ptbhw] [links.6s32f]": IPANT4's "no one I know there" is now Social contact, so this is now being misunderstood or not sharing experience, firm for RVIEW2 and VIEW1M, with VIEW1M's own words quoted.
  • "Loneliness as a cause ... One interviewee says": now IPANT4 (of himself) and RVIEW2 (of people in general).
  • R1: "joins one interviewee's link from loneliness to fear" corrected to IPANT4's and RVIEW2's links.
  • Section heading and count: three reinforcing loops corrected to four (R4 added).
  • R3: "one interviewee only hints at" corrected to two (RVIEW2 and VIEW1M), both weak, each quoted.
  • "Its absence brings loneliness for {loops.c28}": cell renumbered, and MBER21's hint added; quotes now name the speaker.
  • "friends or family whose indifference leaves people feeling ... ({loops.c67}...)": cell renumbered and each quote attributed (YPART2, MBER22).
  • "sisters who make family rifts bearable ({loops.c29})": now says MBER22 is the only firm account of support easing low mood, with VIEW1M's weak hint beside it.
  • "two more only hint that support from services eases loneliness itself": wrong. The two hints come from MBER21, the interviewee already quoted, and VIEW1M, so the sentence now says "he and VIEW1M"; VIEW1M's hint is about services failing rough sleepers, not him, and the report says so.
  • "One interviewee mentions his own counsellor": named RVIEW2 and quoted "I still feel alone".
  • B1: kept, with VIEW1M named; B2 added, said to rest on a weak link.
  • Limits: "asked what loneliness means and where people feel it" corrected to what they associate with loneliness and where they feel most connected and most lonely; "one holds only the second part" now names YPART2.
  • Padding: little to cut. The closing sentence of the old opening ("The first verdict is fairly secure; the second is weaker") was folded into the verdict lines; the limits section was kept, since each point bears on how far the verdicts can be trusted.

The record

This report was drawn directly from the run's zip, do-these-interviewees-describe-loneliness-as-a-2026-10-13-revision-2-b0ccdea2.zip, with Rubicon 0.1.0+247ce85d2434. The zip's fingerprint is b0ccdea2af41dc9d0030552d8006e95f80bbc32f703f9e998f82a170169ee5f5. It is the SHA-256 of the zip's SHA256SUMS file, which lists the SHA-256 of every other file in the zip, so changing any file breaks the match. To check that the zip is the one this report was drawn from, unzip it, run sha256sum -c SHA256SUMS, then sha256sum SHA256SUMS, and compare the result with the fingerprint above. Rubicon's own code, which is public, checks the rest: python verify.py <report> <zip> recounts every number from the coded passages in the zip, draws the report again from the zip and says whether it matches this one exactly.

Revisions

Changes made after the first report, each recounted and checked before this version was drawn. Earlier versions keep their own files.

  • 10 October 2026: Each table now says what it counts: every count names its unit, interviewees, beside the number it is out of. Workflow: every tabulate step names its unit. Coding and answer unchanged. Recounted the whole workflow; no count moved. The recount also states how near each verdict is to moving (its margin), which the engine that made the earlier recount did not.
  • 13 October 2026: Asked to bring the run up to date after a recount with the current version of Rubicon counted one loop differently. Workflow and coding unchanged. The current recount counts a loop as told whole only where all its links are explicit, so B2 (Loneliness, Social contact, Self-worth), which VIEW1M's account closes only through an implicit link, moved from 1 of 6 interviewees telling it whole to 0, with VIEW1M's account listed as implicit. Answer: the sentence on B2 rewritten to say so; the opening conclusion read against the new counts and left unchanged, since both verdicts and the count of balancing loops told whole (B1, 1 of 6) did not move. Recounted the whole workflow; no other count or row moved.

Every interview at a glance

Each row is one interview, each column a reason as the coding defines it. Click a mark to read the passages.

Only implicitly, so left out of the counts
Social contactSupport from family and friendsGoing out and taking partLonelinessSocial confidenceSupport from servicesWork or studySelf-worthBelongingMental wellbeingMoney
Male
Female
🔥 Carbon cost of this report: Between 0.0058 and 1.7 km driven in an average UK family car

Making this report used between 7.7 and 2238 watt-hours of electricity, which is the carbon of driving an average UK family car for that distance. A ballpark only: Anthropic publishes no energy figures, and the run, made in the evaluator's own Claude, records no token counts, so the estimate is made from the run's size. The 6 documents were read in full once by each of the 1 coding steps, by the second coder and by the check, and the writer read the tables; everything the run wrote counts as output. That comes to about 181,544 tokens read and 19,432 written, at 4 characters a token. It covers the first run and its two revisions. The zip holds only the latest state, so each revision is counted at the low end as rereading the tables and writing the answer again, and at the high end as a whole run again.

  • High: 0.39 Wh per thousand tokens read and 1.95 per thousand written, with what was read counted 10 times over for the agent's own overhead. Simon P. Couch, 'Electricity use of AI coding agents', 20 January 2026, https://simonpcouch.com/blog/2026-01-20-cc-impact/: about 390 Wh per million input tokens and 1,950 per million output tokens for Claude Opus, back-solved from Epoch AI's estimate for a 100,000-token prompt with output at five times the input rate, as Anthropic's prices are. The author calls them napkin maths. Our assumption, not a measurement: an agent re-sends its instructions, earlier turns and tool output with every request, so it reads far more than the documents themselves. The low case counts only what was read and written; the high case counts what was read ten times over.
  • Low: the same rates divided by 20, what was read counted once. Oviedo et al., 'Energy use of AI inference, efficiency pathways, and test-time scaling', arXiv:2509.20241: a median 0.31 Wh per query for frontier models, and widely cited estimates overstated by 4 to 20 times. The low case takes the high case's rates divided by the top of that range.
  • 0.124 kg CO2e per kWh. DESNZ greenhouse gas reporting conversion factors, UK grid electricity, 2024.
  • 0.16601 kg CO2e per km. DESNZ greenhouse gas reporting conversion factors, average car of unknown fuel, 2024.

Made with Rubicon, free for Claude: rubicon.causalmap.app