The version of your Co-Parent in your head right now

Stage 1 · The first 90 days · Article 12 · Wave 2

---

In the first weeks of separation, the version of the Co-Parent in your head is not the same as the actual Co-Parent. It's a composite, built from the last terrible argument, the worst weeks of the marriage, the things you discovered or finally let yourself see at the end, plus your own projections about what they're doing now that you can't observe.

This article covers why this gap exists, why it matters for the practical work of co-parenting, the five most common distortions, how to update your internal model without minimising real concerns, and what to do when the actual Co-Parent surprises you.

Why the gap exists

The Co-Parent in your head and the actual Co-Parent are different for a few reasons.

1. You stopped getting new data when the daily contact ended. For years, you had constant updates on who they were, moods, behaviours, small kindnesses, irritations. The separation cut off most of that input. Your model of them is now frozen at the moment the daily contact stopped, which was usually one of the worst phases of the relationship.

2. The brain consolidates the recent worst into the general picture. In the absence of new data, the brain uses the most available data, which is whatever was happening when the relationship was ending. That data is unrepresentative; it's a worst-of compilation rather than a balanced sample.

3. The grief and anger amplify the distortion. Your nervous system has reasons to keep the Co-Parent at a certain distance, emotional protection, anger preservation, justification of the decision. Maintaining a worse-than-real model serves those functions. The protection is useful for a while; the distortion has costs.

4. You're filling in the unknowns. The Co-Parent now has a life you can't see. Most of what they're doing, thinking, and feeling is opaque to you. The brain doesn't tolerate ambiguity well, so it fills the gaps with projection. The projection is usually pessimistic about them and pessimistic about your situation.

The result: the Co-Parent in your head is more difficult than the actual person, especially in the first 90 days.

Why this matters

You might think this gap is just an internal mental artifact, irrelevant to the real practical work. It isn't. The gap has concrete effects.

1. It distorts your reading of their messages. A neutral logistical message from the actual Co-Parent gets filtered through the model of the Co-Parent in your head. The hostile reading wins because the model is primed for hostility. You then reply to the message you predicted, not the one they sent. (See Article 31.)

2. It distorts your decisions. You make choices about logistics, money, the children, communication channels, all of them filtered through the worst-version model. The choices end up calibrated for a worse person than the one you're actually co-parenting with, which produces over-defensive systems and unnecessarily adversarial exchanges.

3. It distorts your child's experience. Your children are reading your responses to the Co-Parent. If your responses are calibrated for the worst version, they pick that up. The children's sense of the Co-Parent is partially shaped by how you respond to them, and you're responding to a composite, not a person.

4. It costs you energy you don't have. Maintaining the worst-version model takes work. The vigilance it requires drains your nervous system in this already-depleted period.

Updating the model isn't about being naive. It isn't about minimising real concerns. It's about responding to the actual Co-Parent rather than the one in your head, which produces better outcomes for you, them, and the children.

The five most common distortions

In the first 90 days, the model tends to distort in five recognisable ways. Spot which ones are running.

Distortion 1: They're worse than they actually are

The most common. You're holding them to the worst version they showed in the late marriage. Behaviours that were rare are now treated as default. Behaviours that fluctuated are now treated as fixed.

Check: can you list three specific positive things about them as a parent or person? If the list takes effort to produce, the distortion is running.

Distortion 2: They're doing better than they actually are

The opposite distortion, less common but real. You assume they're moving on more easily, parenting more confidently, recovering faster. This usually comes from social media posts or filtered observations, the curated version they're presenting publicly.

Check: are you assuming things about their inner state that you have no actual data about? If yes, the distortion is running.

Distortion 3: Their behaviour is more strategic than it actually is

You read their actions as part of a plan. A late pickup is a power move. A short message is a calculated dig. A change in schedule is a manoeuvre.

In reality, most Co-Parent behaviour in this period is also reactive, exhausted, and improvised. They're not running a strategy; they're managing.

Check: when their behaviour seems strategic, ask, what would be the simpler explanation? Usually it's a fitting one.

Distortion 4: They haven't changed at all

Some Co-Parents are slowly changing post-separation. The change might be subtle and not obvious through the channel you have with them. The model in your head doesn't update because there's nothing dramatic to update against.

Check: have you noticed any small shifts in how they communicate, parent, or behave since the separation? If you can't recall noticing anything, you might be missing real updates.

Distortion 5: They've changed entirely

The opposite. Particularly common when you see them after a gap, or hear about them through others. You decide they're a transformed person now, in either direction (better or worse).

Check: is the evidence of transformation specific, repeated, and over time? Or is it a single observation you've extrapolated? Most "transformations" reported at month two turn out to be moods, not changes.

How to update the model without minimising

The goal is accuracy, not generosity. Updating doesn't require forgiving, liking, or trusting. It requires seeing.

A few practices.

Practice 1: Note specific behaviours, not patterns

When you find yourself thinking they always do X, ask: when, specifically? Can you list three actual instances from the last month? If you can't, the pattern is partially constructed.

The brain extrapolates from a few instances to a pattern faster than the data justifies. Catching the extrapolation produces more accurate models.

Practice 2: Distinguish what you've observed from what you've imagined

For each strong belief you hold about the Co-Parent right now, ask: did I see this directly, or am I inferring it?

Most beliefs in this period are a mix. Separating the directly observed from the inferred reduces the inference-creep that builds over weeks.

Practice 3: Keep a brief private record

Not for litigation, not for grievance. For accuracy. A short note each week, what messages were exchanged, how they handled the handover, how the children seemed after returning. Across months, the record reveals patterns that the moment-by-moment perception misses.

The record also shows positive things you would otherwise forget. Most parents who keep one are surprised by how much of the Co-Parent's behaviour is neutral-to-okay when reviewed across weeks.

Practice 4: Let small positive evidence land

The Co-Parent does something competent or kind. The temptation is to dismiss it (they did it for the children, not for me), explain it away, or refuse to count it. Don't.

You don't have to praise them. You don't have to tell them. Just let the data point land internally. Okay, that was handled well. Move on.

Practice 5: Distinguish their behaviour from your feelings about their behaviour

The same action can produce strong feelings in you while being a neutral or even reasonable action by them. Naming the gap reduces the conflation.

Their message was reasonable. My response to it was strong. The strong response is mine to manage; the message isn't the problem.

This isn't dismissing your feelings. It's separating the data about them from the data about you.

What to do when the actual Co-Parent surprises you

Sometimes the actual Co-Parent does something the model in your head didn't predict. They're kind in a moment you expected sharp. They handle something well that you assumed they'd handle badly. They show care for the children that you didn't expect them to show.

A few responses to this.

1. Don't ignore the surprise. The temptation is to file it as anomalous, they're just doing it because of X reason, it doesn't count. This filing preserves the model in your head. Better: notice the surprise. Let it update the model slightly. You don't have to revise everything; you just have to update by one data point.

2. Don't over-update. Some parents go the other way and use a single good moment to retract everything. Maybe they've actually changed, maybe I was wrong about them. This produces whiplash. A single data point doesn't replace a pattern; it adjusts the model by one degree.

3. Don't share the surprise with them. Don't say I was expecting you to handle that worse than you did. This produces a complicated exchange. Internal update; no external communication required.

4. Don't share it with the children either. Your dad/mum is actually being okay this week is not a thing to say to children. Their experience of the Co-Parent is their own. Your internal calibration is your business.

When the model needs not to update

A small but important caveat.

If the Co-Parent's behaviour includes abuse, physical, sexual, financial, sustained psychological, the model should not be updated upward based on intermittent positive behaviour. Abusive Co-Parents often have cycles that include kind or competent moments. Those moments don't mean the underlying pattern has changed.

For these situations, the model in your head being calibrated for vigilance is appropriate. The accuracy work is different and requires specialised support, not a general perspective-taking exercise.

If your situation includes abuse patterns, this article is not the right resource. Please find specialised support.

Quick reference

Five common distortions of the Co-Parent model in early separation:

1. They're worse than they actually are.
2. They're doing better than they actually are.
3. Their behaviour is more strategic than it actually is.
4. They haven't changed at all.
5. They've changed entirely.

Five practices for updating without minimising:

1. Note specific behaviours, not patterns.
2. Distinguish observed from inferred.
3. Keep a brief private record.
4. Let small positive evidence land.
5. Separate their behaviour from your feelings about it.

When the Co-Parent surprises you:

• Don't ignore the surprise.
• Don't over-update.
• Don't share it with them.
• Don't share it with the children.

When not to update:

• Abuse patterns. Vigilance is appropriate. Different work, different support.

The Co-Parent in your head and the actual Co-Parent are different people. Updating the model is for you, not for them.