Chapter Twenty Four: Amplification

The auditorium was the only space in Neurovia designed for an audience.

Everything else on campus minimized the distance between a person and a task. Rooms reconfigured. Corridors adjusted their light. Labs responded automatically to user needs. The auditorium did none of that.

Cohort 2 filed into the third row from the front. Cohort 1 took the rows immediately behind them. No one had told them to sit that way. Julian noticed it only after he had already sat down, the way you notice a door has closed behind you.

Julian let his gaze move across the room. The faculty had arranged themselves in a loose arc behind the podium. Dr. Halvorsen at the center, upright and confident. Dr. Patel to her left, fingers interlaced, eyes moving student to student. Dr. Levin near the edge, weight forward on one foot.
The gap in the formation registered before Julian had located its cause. Dr. Morse was in the back row of the seating rather than on the stage. His folded display was open across his knee, its blue light on his downturned face. His index finger moved periodically across the screen while students continued filing past him. He did not look up.

Dr. Halvorsen stepped to the podium.

* * *

“Three months ago,” she said, “none of you had been here before.”

Her voice carried without amplification.

“You came from different places. Different circumstances. Different understandings of what this program was and what it would ask of you.” She paused. The pause was a deliberate space, not a hesitation. “You have now completed the first phase of Catalyst integration. What that means, beyond the metrics we have been collecting, is something each of you is still discovering. That is appropriate. Integration is not an event. It is a process. And you are still inside it.”

“This week, we begin the second phase.” She moved slightly, her gaze taking in both cohorts without favoring either. “Individual amplification assessment. The purpose is not to rank you or compare you to each other. It is to understand, with precision, how Catalyst interacts with the architecture of each participant. Where it amplifies. Where it meets resistance. And what the ceiling looks like from where you currently stand.”

She let that land before continuing.

“One additional piece of news.” Her tone did not change. “Based on the results of Phase One, Neurovia has received authorization to expand the trial. Two new cohorts, Cohorts 3 and 4, will begin pre-screening this week and implantation within the month.”

The room shifted. No voices. No visible reaction.

“Your data made that possible,” Halvorsen said. “The neural patterns your cohorts have demonstrated, the adaptation curves, the integration metrics, they provided the evidence needed to authorize expansion. I want you to understand that clearly. Not as pressure. As fact. What is happening now shapes the future of this technology, and the future of everyone who will follow you into it.”

Behind Julian, someone in Cohort 1 exhaled. Controlled too late.

“This week's assessments will run concurrently with your standard schedule. You will each receive a personal itinerary this morning. The exercises are built around your individual baseline profiles and the integration data Catalyst has produced over three months.” She looked across both rows. “You may find some of them straightforward. Others will not be. I ask that you approach all of them honestly. Not to perform for us. To show us what is actually there.”

She stepped back from the podium.

“That's all,” she said. “Go find out.”

* * *

The room assigned to Julian was small, clean, and almost featureless.

No projections when he entered. No interface. Four walls, a ceiling, and a floor grid that activated as he stepped inside.

AMPLIFICATION ASSESSMENT — JULIAN REYES

DOMAIN: PREDICTIVE MOTOR RESPONSE / TEMPORAL INTEGRATION

CATALYST STATUS: ACTIVE — FULL MEDIATION

EXTERNAL ASSISTANCE: DISABLED

BASELINE REFERENCE: PRE-INTEGRATION PROFILE LOADED

A single technician sat behind a glass partition at the far end, hands resting on a console.

Julian stood at the center of the grid and waited.

The first stimulus arrived without warning.

A dense foam sphere, fist-sized, dropped from the ceiling at normal gravity. Julian caught it before his eyes had registered it leaving the release point. He stood there with the sphere in his hand, arm still extended at the angle of the catch. He had not told it to do that.

RESPONSE INITIATION: 67ms pre-stimulus

NOTE: Response preceded visual confirmation by 3 frames

The technician made a note without looking up.

The second test released objects in sequence from twenty meters of ceiling track. Spheres, irregular shapes, a flat disc, each weighted differently, each from a different point along the run. The pattern was random, or designed to seem random.

Julian stopped thinking about the objects.

He tracked the release points instead. The fractional shift in a ceiling panel a half-second before each drop. The air pressure differential that preceded each release. He was not catching things anymore. He was occupying positions. His hands moved through the space as part of a continuous loop, extending and closing and redirecting, the gap between perceiving and acting compressed to the point where the distinction stopped mattering.

RESPONSE LATENCY: 11ms average (pre-integration baseline: 189ms)

PREDICTIVE ACCURACY: 94.7% (anticipatory, not reactive)

CATALYST INFERENCE ACTIVITY: EXTREME

NOTE: Subject is not reacting to stimuli. Subject is modeling release parameters.

The third exercise used light. Patterns strobed across the floor grid in escalating frequency. Julian was asked to step on illuminated nodes before they extinguished. At forty nodes per minute it was fast. At sixty, demanding. At eighty, the exercise should have become impossible.

It did not.

Julian moved through the grid the way water moved through a channel it had always known. He was not choosing nodes. He was occupying them. The light was confirmation of decisions already made in some layer of his nervous system that did not have a name yet.

COMPLETION RATE AT 80 NODES/MIN: 100%

COMPLETION RATE AT 120 NODES/MIN: 97.3%

NOTE: Above 80 nodes/min, subject's motor coordination exceeds biomechanical parameters predicted by standard Catalyst integration models.

RECOMMEND: Escalate to secondary assessment tier. Flag for Halvorsen review.

Julian stopped when the exercise stopped. He was breathing harder than he expected. Sweat had gathered at his collar. He became aware, slowly, that his body had been doing something he had not been supervising. Something was using the space inside him more efficiently than he ever had, and he was not sure how he felt about that.

“That's the session, Julian. Thank you.”

He walked to the door. Behind the glass, the technician was typing quickly, not looking up. He noticed that her hands had a slight tremble.

* * *

Theo entered a room of interlocking mechanisms; objects placed with studied informality yet bound by invisible causal chains.

AMPLIFICATION ASSESSMENT — THEO FISCHER

DOMAIN: SYSTEMS COGNITION — EMERGENT PATTERN RESOLUTION

CATALYST STATUS: ACTIVE — FULL MEDIATION

STIMULUS ENVIRONMENT: ADAPTIVE — COMPLEXITY SCALES WITH ENGAGEMENT

A minimal instruction glowed: “Understand it.”

Theo looked at the room for forty-five seconds without touching anything.

The objects were connected. Not obviously, not with visible mechanisms, but causally. The position of a counterweight on the near shelf was in relationship to a magnetic plate on the far table, which was in relationship to a pendulum suspended from the ceiling, which fed into something else he had not located yet. He felt the whole shape of it before he could have said what any individual piece was.

SYSTEM COMPREHENSION: Full architectural model completed in 47 seconds

NOTE: Standard completion benchmark for this configuration is 12-18 minutes.

CATALYST INFERENCE ACTIVITY: EXTREME

He moved to the first intervention point and made a small adjustment to the counterweight. Not a guess. The change the system had been asking for. He watched the cascade propagate. Each response exactly what he had already seen in the model in his head. The system arriving at a new equilibrium more stable than the original.

“Interesting,” he said, to no one in particular.

ADAPTIVE DIFFICULTY: TIER 3 OF 5

Subject reached Tier 3 in 4m 12s. Previous fastest: 23m 08s.

The Tier 4 configuration introduced chaos. Genuine randomness seeded into three of the mechanical relationships, disrupting the causal chains he had been tracking. The system became, briefly, illegible.

Theo sat down on the floor.

Not in defeat. He watched the randomized elements without trying to resolve them, letting the noise accumulate, and looked for what he was actually looking for. The pattern underneath the chaos.

He stood. Identified the three injection points. Resolved the randomness in eleven seconds by addressing how the system compensated for it.

TIER 4 COMPLETE.

Chaos resolution time: 11.3 seconds.

NOTE: Subject did not address random elements directly. Resolved by modeling compensatory responses in adjacent systems. Approach not anticipated in experimental design.

TIER 5 — DIGITAL SYSTEMS INTEGRATION

OBJECTIVE: The monitoring parameters for this session are fixed. Baseline suppression is active. Your results are being recorded at 40% resolution to prevent anomalous data from triggering automatic review.

Override it.

Theo stood very still.

He looked at the terminal. At the words. At the challenge that was also a confession. The system had limits, and someone was telling him exactly where they were, and the question was whether he could move through them.

He crossed the room and sat down in front of the panel.

He did not touch it. Not yet. He sat with his fingers an inch above the keys and let his vision unfocus until the interface blurred at the edges. He was not looking at it. He was looking through it. For the architecture beneath it. Thirty seconds. Sixty.

The terminal was running standard Neurovia monitoring infrastructure. The same backbone that managed climate, access, and data routing across the building. From this instance it controlled only the session. Recording resolution. The alert thresholds that would notify Halvorsen's team if readings crossed certain values. And, he saw this quickly once he knew to look, the data suppression filter they had described. Installed three layers deep inside a routine that looked, from the outside, like ordinary compression.

The disguise was good. Three weeks ago he would not have seen it.

Now it was obvious in the way that foreign words became obvious once you had heard them enough.

He touched the panel for the first time.

He navigated through the system the way he moved through the physical room. Finding the places where the architecture assumed it was not being examined and slipping through them. The access protocol had a handshake sequence. The handshake had a timing vulnerability. The vulnerability opened a gap in the permission structure that lasted, at the right moment, for approximately forty milliseconds.

He waited for the right moment.

Then he was through.

SESSION MONITORING — STATUS UPDATE

[Baseline suppression: DISABLED]

[Recording resolution: 100%]

[Alert thresholds: MODIFIED — all anomalous readings now route to subject display only]

He sat back and read the screen.

He had turned the monitoring system around. What had been reporting to the faculty was now reporting to him. The alert thresholds that would flag results too unusual for standard review now appeared on his terminal instead of Halvorsen's. The supervisor behind the glass was looking at a console that showed a clean, unremarkable session. Everything within expected ranges. Nothing to flag.

None of that was true.

Her system no longer knew it.

TIER 5 COMPLETE.

NOTE: Subject accessed, modified, and redirected session monitoring infrastructure without triggering a single access alert.

NOTE: Supervisor console currently displaying falsified session data. Duration of falsification: ongoing.

NOTE: This note is visible only on subject's terminal.

NOTE: Recommend immediate escalation. Assessment team is presently unaware of Tier 5 completion.

Theo read the notes.

Then he restored the original monitoring parameters. All of them, cleanly, leaving no trace he could identify.

* * *

Mateo waited for his briefing longer others. Eventually the technician said: “We are going to run a decision cascade. There is no time limit, but the scenarios will respond to your choices in real time.” A pause. “Do not try to optimize. Just decide.”

Mateo thought: those are not compatible instructions. He did not say so.

AMPLIFICATION ASSESSMENT — MATEO DELUCA

DOMAIN: PROBABILISTIC DECISION ARCHITECTURE — OUTCOME CONVERGENCE

CATALYST STATUS: ACTIVE — FULL MEDIATION

NOTE: Participant flagged for emergent behavior observation based on prior session data.

The first scenario was a logistics problem. Routes, capacities, timing windows. Mateo solved it in forty seconds.

The second was a social system. Two actors, competing interests, imperfect information distributed unevenly between them. He worked more slowly here, understanding that getting it right mattered more than getting it fast.

He got it right.

SCENARIO 2 ACCURACY: 100%

NOTE: Subject correctly modeled three hidden actor relationships not included in the briefing material.

The third scenario had no solution.

He knew immediately. The way every path led to a state that generated the conditions for the next failure. It was a system with no stable equilibrium. It was not asking him to decide. It was asking him to recognize that deciding was the wrong answer.

He said so.

A pause from behind the glass. Then: “Can you be more specific.”

“The scenario assumes all actors will update their behavior in response to new information. Actor four will not. Not because they lack the capacity. Updating would require acknowledging that a prior decision was wrong, and their pride will not allow that. Remove that actor and the system has seventeen stable solutions. With it, it has none.”

SCENARIO 3: No solution by design.

Subject identified structural flaw in scenario architecture. Correct.

NOTE: Subject completed meta-analysis of problem set rather than engaging with problem as presented.

The fourth scenario was a room. Rendered, populated, dynamic. The projections moving and responding in real time. He was told to watch for two minutes, then describe what would happen next.

He watched.

The familiar compression arrived. The narrowing of futures into a single clear channel. The branching possibilities consolidating as he watched the people move through the space. Easier than it had been during Ghost.

“The woman at the left will leave in approximately ninety seconds,” he said. “The man beside her will use the absence to say something he has been waiting to say. He has been holding it since before this scene started. The group will fracture along the tension that has been running between the two at the center, which has been waiting for a catalyst.” He paused. “The catalyst is the woman leaving.”

The woman left at eighty-three seconds.

PREDICTIVE ACCURACY: 94%

CATALYST CONVERGENCE ACTIVITY: EXTREME

NOTE: Preliminary analysis suggests subject's stated predictions may be partially self-fulfilling. Actor behavioral parameters shift in response to subject's gaze and posture, not only scenario programming.

FLAG: Notify Morse and Patel immediately. Do not inform subject.

* * *

Leena's assessment room looked, at first, completely normal.

That was the first thing she noticed was wrong.

Every other room she had been in at Neurovia was designed for something. Its purpose apparent in the choices that had made it. This room had a table, four chairs, a window overlooking a courtyard, and nothing else. It was the most ordinary space she had encountered in three months, and because everything at Neurovia was intentional, that ordinariness was itself a design choice.

She stood in the doorway and looked at it for a full minute before stepping inside.

AMPLIFICATION ASSESSMENT — LEENA PARK

DOMAIN: MULTI-LAYERED PERCEPTION — DIMENSIONAL SIGNAL INTEGRATION

CATALYST STATUS: ACTIVE — FULL MEDIATION

ENVIRONMENT NOTE: No stimuli have been introduced. Room is baseline standard.

The technician, a woman Leena did not recognize, young, sitting at the table, said: “Describe what you see.”

“The room.”

“Yes.”

Leena looked at it again.

She felt Catalyst shift. Not with any sensation she could locate, more like adjusting to a new light level.

“The window frame,” she said slowly, “has been painted three times. The most recent layer is about a month old. The layer beneath it has a stress fracture from when the building settled. I can see it through the surface paint as a difference in how the light reflects.”

The technician wrote something.

“The table has been moved recently. The legs left marks in the floor composite. They are not sitting in the marks. Someone moved it two or three meters and back again, maybe a week ago.” She tilted her head. “The chair you are in has been sat in by at least four different people today. I can read it in the pressure distribution pattern in the seat material.”

The technician looked down at her chair. Then back up at Leena.

“Continue.” Her voice changed slightly.

PERCEPTION DEPTH: Exceeding standard sensory acuity. Structural analysis of room materials ongoing.

NOTE: Subject is reading surface histories, stress patterns, and use traces from standard environmental visual input. Not magnification.

Leena walked to the window.

The courtyard outside was empty. Flagstones, a dormant planter, a bench. Not quiet to her. She could read where water had pooled and evaporated in patterns that told her something about the drainage beneath the stones. She could see each flagstone's thermal signature still resolving from the afternoon sun. Each one at a slightly different temperature based on its density and position and how long it had been in shade. She could see, in the bench's wood grain, the particular warping that told her it faced west and had been doing so for at least eight years.

“There is something wrong with the fourth flagstone from the left,” she said. “It sits two millimeters lower than its neighbors. Whatever is underneath has shifted.” She paused. “It happened recently. The water drainage pattern around it has not fully adjusted yet.”

The technician put her pen down. Picked it up again.

“We will move to the second phase. Please sit down.”

NOTE: Flagstone observation forwarded to facilities. Confirmed: subsurface pipe joint displaced 2.1mm. Maintenance not previously aware.

NOTE: Subject identified structural anomaly invisible to standard inspection.

The second phase was sound.

A speaker in the ceiling produced a single tone. Middle C, clean, one second long. They asked her to describe it.

Leena sat with it.

The note was not a single thing. She understood this now in a way she could not have three months ago. She could hear each frequency and harmonic. She could hear the speaker’s own resonant frequency underneath the tone it was producing. A ghost vibrating in sympathy with itself.
She could hear the room respond. The table’s wood dampening the upper frequencies. The painted walls sharpening the midrange. The window glass giving back something cooler and thinner.
She could hear, beneath all of that, the building. Ventilation, electrical current, water in distant pipes, each briefly disturbed by the tone and already settling back.

“It is not one sound,” she said. “It is.” She stopped. There was no word for what she was reaching toward. “It is a location. A point where a lot of different things are happening simultaneously, and the note is what they look like from the outside.”

The technician stared at her.

“I know that does not quite make sense.”

“No,” the technician said slowly. “It does.”

AUDITORY PERCEPTION: Full harmonic decomposition. Active.

NOTE: Subject is resolving acoustic signals into component frequencies, material interactions, and spatial resonance simultaneously.

The next phase was the one that made the technician go very still.

She placed a sealed box on the table. Metal, roughly the size of a hardback book, no visible seam or opening.

“Tell me what is inside.”

Leena looked at it. She was not sure what she was looking for. She looked anyway, letting information arrive rather than seeking it.

The weight was distributed unevenly toward the bottom-left. Whatever was inside was dense but small, not filling the space. The metal itself was warmer than ambient on one face. Either the object inside was generating negligible heat or it had recently been in contact with something warmer. She picked the box up and felt the object settle fractionally. Not rolling. An irregular surface. She set it down and looked at the face where the warmth was greater.

“It is a key,” she said. “Old-fashioned, physical. Brass or something similar. A material with higher specific heat than steel. It has been handled recently. Within the last two hours.”

The technician opened the box.

A key. Brass. Old.

She set the box down with the care of someone whose hands were not entirely trustworthy in that moment.

PHASE 3 RESULT: Correct. Object, material, recent handling. All accurate.

NOTE: Object identification achieved through thermal differential reading, weight distribution analysis, and micro-vibration response during handling. Subject is integrating multiple sensory inputs into a unified cross-dimensional model.

* * *

Ethan's assessment was the most straightforward in design.

The room was large and padded on all surfaces. Equipment distributed across the space: resistance bands, weighted bars, a climbing structure along one wall, a central platform inset with haptic sensors that would register force, speed, and precision simultaneously. He looked at it and felt something uncomplicate in his chest. This was the one kind of room where the question was simple, even when the answer was not.

AMPLIFICATION ASSESSMENT — ETHAN CROSS

DOMAIN: PHYSICAL AMPLIFICATION — CONTROLLED OUTPUT CALIBRATION

CATALYST STATUS: ACTIVE — FULL MEDIATION

INHIBITORY PROTOCOLS: PRESENT

NOTE: Ceiling raised per post-Phase-One adjustment

The first sequence was measurement. Grip strength, jump threshold, reaction time, sustained output. He held back instinctively.

The technician's voice came through the speaker after the first sequence: “You can push harder.”

Ethan looked at the sensor platform. “How much harder.”

“We have adjusted the ceiling. Find out.”

He found out.

Carefully, the way you approached an edge in the dark. One foot. Then weight. Then the next foot. Then weight again. Each increment deliberate. Building a picture of where the edge actually was rather than where you had assumed it would be. He pushed against the resistance systems and felt them give more than expected. Pushed harder. Felt them recalibrate to match him. Pushed again.

At some point the holding back stopped.

He climbed the wall structure at a pace the technician would watch on footage three times before logging it. He hit the platform sensors at forces the haptic system was not rated for. He caught a weighted bar dropped from eight feet without bracing.

OUTPUT METRICS: EXCEEDING TIER 4 PARAMETERS

PEAK FORCE OUTPUT: 340% above pre-integration baseline

REACTION TIME (PHYSICAL): 8ms. Below documented human threshold.

NOTE: Inhibitory protocols functioning. Subject is in controlled output that exceeds what our control parameters were designed to measure.

Ethan stopped when the equipment started to break.

He stood in the center of the room, breathing hard, hands loose at his sides. The spent feeling was good. Underneath the good, something else. A wrongness he could not place. A sense of having crossed a line he had spent his whole life carefully staying behind.

* * *

Zara Osei had been at Neurovia for eighty-seven days, two hours, and fifty-four minutes, and in all that time she had spoken, by her own precise count, three hundred and ninety-four words to the other members of Cohort 2.

The others had noticed. They accepted her as she is without pushing, which she appreciated more than she could have said without exceeding her word count significantly. She had seen Julian watch her with a furrowed brow, working out whether to reach in. Theo had once slid a note across the table that read “You OK?” which she had folded into precise quarters and returned without response. Not because she was not okay. The response she wanted to give, the one that was true, had too many moving parts, and the translation would have failed.

AMPLIFICATION ASSESSMENT — ZARA OSEI

DOMAIN: DIRECT NEURAL COMMUNICATION WITH CATALYST-MEDIATED SYSTEM INTERFACE

CATALYST STATUS: ACTIVE — FULL MEDIATION

ENVIRONMENT: SPECIALIZED — CATALYST-EQUIPPED CONSTRUCT UNITS PRESENT

The room was larger than she had expected. It was not empty.

Three units stood along the far wall. Roughly human-proportioned.

The technician said: “You do not need to do anything in particular. Just be aware of them.”

Zara looked at the units.

She felt something she recognized before she had named it. Like noticing a conversation already in progress that she had not heard begin. The units had digital systems. Catalyst had, over three months, become a language she thought in as fluently as she spoke in any other. The units were broadcasting on it. Not intelligently, but the signal was there. An open line. The equivalent of a phone left off the hook in an empty room, waiting for someone to notice.

She picked it up.

ZARA OSEI — NEURAL INTERFACE ACTIVITY: DETECTED

NOTE: Subject has established Catalyst-mediated signal connection with Unit 1. Method: unknown. No instruction provided.

CONNECTION LATENCY: 0.3ms

Unit 1 moved. Just its head. Turning slightly toward her. It turned toward her the way a person turned toward the person they were listening to.

Zara was not sure how to describe what she was doing. Intention translated into instruction. She thought about the unit turning toward the door.

Unit 1 turned toward the door.

“Can you,” the technician started.

“Please do not talk for a moment,” Zara said.

The technician stopped.

Zara reached for Unit 2. The connection was easier now. She knew the shape of the signal. The frequency. She held it alongside the first.

CONCURRENT CONNECTIONS: 2

NOTE: Subject is maintaining simultaneous neural interface with two distinct Catalyst-equipped units. NOTE: No degradation in connection quality for either unit.

She reached for Unit 3.

This one was different. Something in its calibration was off from the others. A different frequency. A different handshake. She adjusted for it without deciding to. The way you matched pitch.

All three units moved simultaneously. Turning. Orienting. Arranging themselves in a loose triangle with Zara at the center.

CONCURRENT CONNECTIONS: 3 (full available set)

RESPONSE FIDELITY: 99.7%

NOTE: Subject has integrated all three units into a single coherent mental model. Units are responding to intent, not instruction.

The technician had stopped writing. She was watching.

The second phase was practical. A series of physical tasks. Objects to move. Configurations to achieve. A course the units needed to navigate. Standard tests designed for controlled remote operation through a conventional interface.

Zara sat still, hands open on her knees, eyes open, and the units moved through the tasks with a fluency that did not look like operation.

TASK COMPLETION: 4m 11s (estimated: 18m)

ERROR RATE: 0.0%

NOTE: Units operated with complete spatial awareness of each other. No collisions, no redundant actions, no coordination lag. Subject appears to be modeling all three units' spatial positions simultaneously as a unified field.

When the tasks completed, Zara released the connections slowly. She let each connection close in sequence, and the room became ordinary-sized again.

The units returned to their positions at the wall.

“That's remarkable,” the technician said.

Zara nodded. She agreed. What she was thinking about was the third unit. The one with the slightly different frequency, the different handshake. The way she had adjusted for it without deciding to. The way she had known, without being told or working it out, what the right frequency was.

* * *

The notes in Kael Draven's file used careful language.

Dissociative presentation. Adaptive identity expression. Context-responsive cognitive modality shifts.

What they meant was: Kael was not always Kael.

Kael had not been fully aware of this before Neurovia. He had registered episodes of lost time. He had noticed handwriting in his notebooks he did not remember producing. His parents had noticed too, and had taken him to a series of professionals who had used different words and arrived at different working hypotheses, none of which had quite resolved into a diagnosis his family was ready to accept. Then Neurovia had selected him, and the intake assessment had named what nobody had named to him before, and he had spent three months learning to recognize the others as residents who came and went depending on what the building needed. The arrangement had been working in its imperfect way since he had begun to acknowledge it.

When Kael himself was present, the baseline, the one who answered to the name, he moved slowly, spoke carefully, gave the impression conserving something. The others were less conservative. That was, broadly, the arrangement.

AMPLIFICATION ASSESSMENT — KAEL DRAVEN

DOMAIN: ADAPTIVE IDENTITY ARCHITECTURE — DISTRIBUTED COGNITIVE MODALITY

CATALYST STATUS: ACTIVE — FULL MEDIATION

ENVIRONMENT: MULTI-PHASE — STIMULUS DESIGNED TO TRIGGER MODALITY TRANSITIONS

CLINICAL NOTE: Assessment team briefed on participant's presentation. Do not address apparent identity transitions directly. Observe.

The first phase was mathematics.

Pure, clean, fast. Sequences and proofs that escalated in difficulty. Kael sat at the table and worked through the first three with careful methodical attention.

By the fourth problem, the pace had become uncomfortable.

The specific kind of pressure that, in Kael's experience, was a form of invitation.

He felt a shift in his body, a loosening. A gear change in a vehicle that had been laboring in the wrong ratio.

He sat up straighter. His posture changed first. Then his hands. Where Kael held a pen loosely, this one gripped with precision. The handwriting was different. Smaller. Faster.

The fourth problem solved in forty seconds.

The fifth in twenty-three.

The sixth, a multi-variable proof that had been estimated to take fifteen minutes, in four.

MODALITY TRANSITION DETECTED — 14:23:07

Baseline cognitive profile: SHIFTED

Mathematical processing speed: +470% from Kael-baseline

Catalyst integration: maintaining continuity across transition

CONFIRM: Second modality. Designating M2.

M2 completed the mathematical sequence in its entirety, set the pen down with a small precise motion, and folded his hands on the table.

The second phase began. The room changed.

The lights dropped. The temperature fell. The display panels filled with fast-moving imagery. Conflict. Urgency. A tactical environment unfolding in real time. The atmosphere compressed into something that demanded fast physical and strategic response rather than analytical precision.

M2 did not like this.

The boy at the table shifted. The precise posture loosened. The small folded hands opened and moved to the table's edge. A pause of three seconds.

Then the chair scraped back.

He stood differently. Where Kael was watchful and M2 had been precise, this one read the room as terrain. His shoulders settled lower. His chin lifted. His eyes stopped reading and started scanning.

SECOND MODALITY TRANSITION DETECTED — 14:31:44

Presenting modality: THIRD (designating M3)

Physical readiness indicators: ELEVATED

NOTE: Catalyst motor coordination pathways shifting to match new modality. Transition complete in 1.8 seconds. This is faster than documented pharmaceutical identity transitions by a factor of 40.

The simulation was a complex tactical environment. Multiple moving variables. Decisions with immediate physical consequences in the rendered space. M3 moved through it as though it was natural though the records showed no formal tactical training in Kael's file.

Then the simulation presented a problem that required analytical resolution before physical action was possible. A locked sequence. Data first, then movement.

M3 stopped.

This was not M3's problem to solve, and M3 knew that with the same certainty you had in the dark when you reached for a step that was not there. The knowledge was physical before it was cognitive.

The chair scraped.

He sat back down. The shoulders dropped. The chin lowered. The scanning eyes returned to reading.

M2 worked the lock in ninety seconds.

The chair scraped again.

M3 finished the sequence.

PHASE 2 COMPLETE

OUTCOME: Optimal. All objectives achieved.

TRANSITION COUNT (PHASE 2): 2

NOTE: Subject transitioned between modalities mid-task in response to problem-type requirements. Each modality operated within its own distinct capability set. No blending or overlap detected.

NOTE: Subject's ability to recognize which modality a task requires, and yield to it, appears to be the primary amplified capability. This is distinct from the modalities themselves.

FLAG: Patel — priority review. Also note: a fourth modality signature was briefly detected during Phase 2 transition window. Duration 0.4 seconds. Did not fully surface. Logging for longitudinal tracking.

The lights came back up. The temperature normalized.

There was a long pause.

The chair scraped again. A smaller scrape this time, quieter. Kael sat back down. The original one. The watchful, careful one who held things loosely and moved as though conserving something.

“Did it work.”

“Yes,” the technician said.

Kael nodded slowly.

* * *

The monitoring suite had assessment feeds completing across multiple displays. Flags populating the review queue faster than they were being cleared. The continuous Catalyst telemetry running its uninterrupted thread through all of it. Not pausing for end-of-session markers. Not distinguishing between what the faculty had been watching for and what had arrived instead.

Dr. Patel stood at the back wall with her arms folded, reading. She had not sat down since the data began coming in.

Dr. Halvorsen was at the central console, working through the flagged items in order. There were more flags than the system had generated across the entire first phase of the trial.

Adisa, a junior researcher, twenty-eight, four months at Neurovia, in none of those months having found occasion to feel genuinely uncertain about what he was looking at, cleared his throat.

“The Cross output metrics. I want to confirm I am reading these correctly.”

“You are,” Halvorsen said, without looking up.

“And the Park flagstone.”

“Facilities confirmed it an hour ago.”

Adisa closed his mouth. Then, after a moment, opened it again.

“Is Dr. Morse joining us.”

Halvorsen did not look up.

“He is reviewing the data through his own channel,” she said. “We will reconcile later.”

Dr. Levin pulled up the Reyes feed and ran it again from the node-tracking sequence. He watched Julian move through the grid. The way each contact point was already occupied before the light appeared rather than after. He watched it twice. Then he said, carefully: “Pre-cognition.”

“Predictive motor synthesis,” Halvorsen said. “More accurate terminology.”

“The result is the same either way.”

“The result is the same. The mechanism matters.” She said it quietly, in the tone she used when she was talking to herself as much as to the room.

Dr. Patel set her display down on the table beside her.

“The Fischer sequence. He did not just complete Tier 5.”

Halvorsen nodded.

“He redirected the monitoring feed to his own terminal. Then he restored it. There is a four-minute window during which the supervisor's console showed falsified data, and the supervisor was present in the room and did not know.”

The room held that.

“He restored it voluntarily,” Levin said.

“Yes. He did. That is not the point. The point is that he could do it. The point is that Catalyst showed him how without being asked to show him. The point is that we have a teenager who has just demonstrated the ability to modify institutional monitoring infrastructure without triggering a single access alert, and we are going to give him harder tasks tomorrow.”

No one responded to that.

Halvorsen advanced to the Osei feed. The three units moving with the practiced bonded ease of something that had always been integrated. No coordination lag. No collision. No redundant motion. The course completing in a fraction of its projected time. Zara at the center, hands open on her knees, not appearing to do anything at all.

“Three simultaneous connections,” Adisa said. “99.7% fidelity.”

“She was not controlling them,” Halvorsen said. “She was.” She stopped.

“Thinking through them,” Patel said quietly.

Halvorsen did not respond to that either. She advanced to the Draven footage.

The room watched the second transition in silence. M3 emerging from M2's stillness with the chair scrape and the shift in focal distance. Then the clean reversal when the task required it. Two complete transitions in under four minutes. Each one distinct. Each one fully inhabited, not partial.

“The fourth modality signature,” Patel said. “The 0.4 second detection.”

“We are logging it,” Halvorsen said.

“He does not know it is there.”

“No.” Halvorsen's expression remained neutral. “Not yet.”

“The primary capability with Draven is not the modalities themselves. It is the recognition. Knowing which one the situation requires, and getting out of its way. That capacity is not in the modalities. It is in him. It is Kael.”

Levin considered this. “That is not a symptom. That is a skill.”

“It is both. That is what makes it complicated.”

The room was quiet for a long moment.

No one looked at the empty chair where Morse would have sat.

No one mentioned it again.

Eventually Halvorsen reached the end of the flagged queue. She looked at the array of displays. Eight portraits. Eight neural maps. Eight sets of numbers that were not the numbers any of them had built the prediction models around. She sat with the array for a moment.

“Tomorrow,” she said, “we give them harder tasks.”

Patel looked at her. “Because we need better data.”

“Because we need to know where it stops,” Halvorsen said.

She gathered her display and left. Her footsteps in the corridor, even, unhurried, faded and were gone.

Patel stayed.

She stood alone in the monitoring suite as the feeds cycled softly around her, and thought about what she had written that morning in her private document before coming in. The two lines she had underlined. The thing she had titled OBSERVATIONS — INDEPENDENT and had not shared with anyone.

She had written, among other things, that she did not yet know what Catalyst was becoming.

She had a better answer now. She did not find it comforting.

* * *

They found each other by instinct that evening.

Julian arrived at the common room first and sat without turning on more lights. Then Leena. Then the others in ones and twos, each arriving as though pulled by the same current from different directions. Zara came last. She sat at the edge of the group, hands open on her knees, which was simply how she sat.

Nobody had eaten much at dinner.

“So,” Mateo said.

A long pause. Not an uncomfortable one. The kind of pause that meant everyone was in the same place about what had happened and none of them had words for it yet.

“Yeah,” Ethan said.

Julian turned his water glass slowly in his hands. He did not have language for what had happened in his session. The feeling that the part of him that noticed and the part that acted had become the same part.

“They are going to give us harder tasks tomorrow,” Aisha said.

Ben looked over. “You know that.”

“I inferred it from how the technicians left. The way people leave a room when the data surprised them and they have not processed it yet.”

Zara spoke for the first time all evening. “The units are still in the room,” she said quietly. “I can feel the carrier signal from here.”

Everyone looked at her.

“It is faint,” she added. “It is there.”

The room was quiet.

Julian looked around at all of them. Some he had known for months. Some he still barely knew. Each of whom had spent today alone in a room finding something, or losing something, or both, and could not yet tell the difference.

He stood. Pushed his chair in.

“We're still us,” he said. He was not sure if he was saying it for them or for himself. “Whatever the numbers say. Whatever they found in those rooms today. We're still.”

He stopped.

He could not finish it.

Not because it was not true.

He was not sure, anymore, exactly what us was going to mean going forward. What it meant to be yourself when the self in question was still in the process of arriving. He had thought, when he came here, that he was becoming something. He was beginning to understand that the becoming did not have a destination he could see from here.