What Makes a Hook Score 90+?

BrainScore

April 3, 2026 · 8 min read

When you use Cook to generate hooks, each one comes with a BrainScore. It's a number from 0 to 100 that predicts how strongly that hook will grab and hold attention in short-form video.

But it's not a random number. And it's not based on "vibes."

BrainScore is built on seven neuroscience dimensions derived from Meta's TRIBE v2 research, a foundation model trained on over 1,100 hours of fMRI brain-imaging data from 720 subjects. That research mapped exactly what happens in the human brain when it encounters different types of content, and Cook uses those findings to score your hooks before you ever film them.

Here's what each dimension measures, why it matters for short-form video, and what separates a high-scoring hook from a low one.

1. Attention

What it measures: How likely the hook is to capture immediate, involuntary attention in the first 1 to 2 seconds.

This is the gatekeeper dimension. If Attention scores low, nothing else matters because nobody's watching long enough to experience the other six dimensions.

Attention in neuroscience isn't about being loud or flashy. It's about novelty and prediction error. Your brain is constantly predicting what comes next. When something violates that prediction, attention spikes. That's the neurological basis of the pattern interrupt.

High-scoring: "Stop using your retinol like that. You're making it worse."
Low-scoring: "Here's my nighttime skincare routine featuring my favorite retinol."

The first hook creates prediction error. You assumed retinol was good. Now someone's telling you you're using it wrong. The second hook is exactly what you'd expect from a skincare video. No surprise, no spike, no reason to stop scrolling.

2. Emotion

What it measures: The intensity of emotional response the hook generates, regardless of whether that emotion is positive or negative.

Emotion is what transforms a viewer from passive scroller to active watcher. When content triggers an emotional response, the amygdala lights up and signals to the rest of the brain: "Pay attention. This matters."

The TRIBE v2 research showed that content generating strong emotional responses had significantly higher completion rates. Not because people enjoyed it more (sometimes the emotion was negative), but because emotional engagement creates a biological need to see what happens next.

High-scoring: "My skin was so bad last year that I canceled a job interview because I couldn't cover it up."
Low-scoring: "This product really helped improve my skin texture over time."

The first hook makes you feel something. Empathy, recognition, maybe even a memory of your own. The second hook is informative but emotionally flat.

3. Cognitive Load

What it measures: How easy the hook is to process in the first 2 seconds. Lower cognitive load means higher score.

This one is counterintuitive. You might think a more complex, clever hook would perform better. The research says the opposite.

When cognitive load is too high, meaning the viewer has to work too hard to understand what's happening, the brain's default response is to disengage. On TikTok, that means scroll away. You have roughly 2 seconds, and if the viewer spends those 2 seconds confused, you've lost them.

This doesn't mean your hooks should be dumbed down. It means they should be instantly clear. One idea. One thread. No setup required.

High-scoring: "I replaced my entire morning routine with one product."
Low-scoring: "After researching the pH levels of various cleansers and comparing them to my skin's natural acid mantle, I discovered something interesting."

The first hook is one clean idea you can grasp in under a second. The second requires too much processing for a cold viewer who doesn't know you yet.

4. Reward

What it measures: How strongly the hook activates the brain's reward anticipation circuitry. Basically, does the viewer expect that watching this video will be "worth it"?

The nucleus accumbens, the brain's reward center, doesn't just light up when you receive a reward. It lights up when you anticipate one. A high-Reward hook promises the viewer something valuable is coming: a secret, a transformation, a hack, a reveal.

This is closely tied to the curiosity gap, but it goes beyond curiosity. The viewer needs to believe that the payoff will be worth their time.

High-scoring: "This $8 product outperformed my $90 serum and I have the before-and-after to prove it."
Low-scoring: "I've been trying a new affordable serum for a few weeks."

The first hook promises a specific, concrete reward: a dramatic comparison with proof. The second hook doesn't promise anything. You might keep watching, but there's no reward pulling you forward.

5. Authenticity

What it measures: How much the hook sounds like genuine human speech versus scripted advertising.

The TRIBE v2 research revealed something that every TikTok creator already knows intuitively: the brain responds differently to content it perceives as authentic versus content it perceives as performative. Authentic content activates trust-related neural pathways. Performative content activates skepticism.

On TikTok specifically, this dimension matters more than on any other platform. The entire culture is built on rawness and realness. A hook that sounds like it was written by a brand's marketing team gets scrolled past, even if the information is good.

High-scoring: "Okay so I wasn't going to post this but I literally gasped when I took my makeup off last night."
Low-scoring: "Introducing my new favorite skincare product that delivers visible results."

The first hook sounds like a real person talking to their phone. The second sounds like a product description. On TikTok, the first one wins every time.

6. Watch Curve

What it measures: The predicted retention pattern. Will viewers stay through the full video, or will they drop off after a few seconds?

Watch Curve is about momentum. A hook can grab attention in the first second but fail to create enough forward pull to keep the viewer watching. The best hooks create what the TRIBE v2 researchers call "narrative tension," a sense that the most important information is still coming.

Think of Watch Curve as the difference between a firecracker (one big bang, then nothing) and a fuse (a spark that leads to something bigger).

High-scoring: "Three months ago my skin was at its worst. This morning a stranger stopped me to ask about my routine."
Low-scoring: "I got clear skin using these three products."

The first hook sets up a journey. There's a gap between "worst" and "stranger complimenting me" that the viewer needs to see filled. The second hook gives away the structure upfront, so there's less reason to keep watching.

7. Valence

What it measures: The emotional polarity of the hook, whether it leans positive, negative, or creates a shift between the two.

Valence is subtle but powerful. The TRIBE v2 research found that the most engaging content often creates a valence shift, starting negative and moving positive, or starting positive and introducing tension.

Pure positive hooks ("I love this product so much!") tend to score moderate because they don't create tension. Pure negative hooks ("This product is terrible") can score high on Attention but low on Reward. The sweet spot is a hook that contains both: a negative starting point with the promise of a positive resolution.

High-scoring: "I almost threw this away after the first use. Two weeks later it's the only thing I reach for."
Low-scoring: "I really love this moisturizer, it's so hydrating."

The valence shift in the first hook creates narrative tension. The viewer needs to know what changed between "almost threw it away" and "only thing I reach for." That shift is where the magic happens.

So what makes a hook score 90+?

A 90+ BrainScore doesn't mean a hook is perfect on every dimension. It means the hook performs exceptionally well across the combination of all seven.

In practice, a 90+ hook usually nails Attention and Emotion first (because without those, nothing else gets activated), while also keeping Cognitive Load low and creating strong Reward anticipation. Authenticity tends to be the tiebreaker between a hook in the 80s and a hook in the 90s.

The hooks that score highest tend to share these qualities:

They lead with a specific, concrete detail (not a generic claim). They create a gap the viewer needs to close. They sound like a real person, not a script. And they contain movement, either a valence shift or a narrative arc, that pulls the viewer forward.

The beauty of having these seven dimensions scored individually is that you can diagnose exactly where a hook is falling short. A hook scoring 85 might just need a stronger opening line to boost Attention, or a more specific detail to increase Reward anticipation. You're not guessing anymore. You're optimizing with data.

The research behind it

If you want to dig deeper into the neuroscience, Meta's TRIBE v2 research is publicly available. It's the largest brain-predictive foundation model ever built for understanding how humans process media content, trained on fMRI data from 720 subjects watching over 1,100 hours of video.

Cook is the only UGC tool that applies this research to hook scoring. Most other tools generate hooks based on what looks good on paper. Cook scores them based on what the human brain actually responds to.

That's a different thing entirely. And if you're a creator who wants to know your hook will land before you spend an hour filming, delivering, and editing, it's the kind of difference that changes your workflow.