The goal of this clinical trial is to learn if virtual reality (VR) helps improve thinking and memory skills in adults who have stayed in the intensive care unit (ICU). The study focuses on people who needed a breathing machine or stayed in the ICU for several days and are at risk for memory or "brain fog" issues.

The main questions it aims to answer are:

• Does using VR improve a participant's memory, attention, and thinking skills after an ICU stay?
• Does the "immersive" feel of a VR headset work better to improve these skills than using a handheld tablet?

Researchers will compare three groups to see how different types of care affect the brain:

• VR-Rehab: Participants use a VR headset to play brain-training games.
• Tablet-Rehab: Participants use a handheld tablet to play the same brain-training games.
• Standard Care: Participants receive the usual hospital care without digital brain games.

Participants will:

• Play brain-training games for 12 minutes every day for up to one week while in the hospital.
• Complete memory and thinking tests with a researcher at the start of the study and again after two weeks.
• Answer follow-up questions about their memory and thinking skills for 6 months after leaving the hospital.

Study Rationale and the Immersion Hypothesis Survivors of critical illness often face significant cognitive impairments, a key component of Post-Intensive Care Syndrome (PICS). While early rehabilitation is recommended, traditional methods often struggle to address cognitive recovery in the acute ICU setting.

This trial is designed to investigate the "Immersion Hypothesis." This hypothesis suggests that the neurological engagement provided by a fully immersive virtual reality (VR) environment offers superior cognitive benefits compared to standard digital content delivered via a 2D interface. By utilizing a three-arm randomized design, the study seeks to isolate whether the therapeutic value comes from the digital exercises themselves or the physiological and psychological impact of total immersion.

Study Design and Arms

Participants are randomized into one of three parallel groups with a 1:1:1 allocation ratio:

1. VR-Based Rehabilitation (VR-Rehab): Participants use a Meta Quest Pro head-mounted display (HMD) to engage in the Enhance VR platform (Virtuleap). This involves 6 specific games targeting information processing, problem-solving, attention, memory, and motor control.
2. Tablet-Based Rehabilitation (Tablet-Rehab): Participants engage in the exact same cognitive exercises and gameplay durations as the VR group, but via a 2D tablet interface. This arm serves as a "digital control" to account for the effects of cognitive stimulation without immersion.
3. Standard of Care (SOC): Participants receive conventional ICU rehabilitation, including physical and respiratory physiotherapy, without structured digital cognitive training.

Intervention Protocol All digital interventions are performed at the bedside with the participant positioned at a 30° to 60° angle for safety. Sessions are designed for a target "dose" of 12 minutes of active gameplay per day for up to 7 consecutive days (or until ICU discharge).

A "Minimum Therapeutic Dose" is defined for this study as the successful completion of at least 12 individual game sessions. This threshold ensures that participants have reached a level of engagement sufficient to evaluate the impact of the technology on cognitive recovery.

Procedural Rigor and Bias Mitigation

To ensure high-quality data in a non-blinded clinical environment, the trial enforces a strict separation of duties:

• Researcher Roles: Different researchers facilitate the daily intervention sessions and conduct the primary cognitive assessments. Those delivering the VR or Tablet therapy are prohibited from scoring the recovery outcomes.
• Automated Analysis: To eliminate subjective interpretation, primary group comparisons are performed using pre-developed, automated scripts in Python.
• Sanitization: Strict infection control protocols are followed, using medical-grade, non-corrosive disinfectant wipes for all hardware between uses.

Safety Monitoring Participant safety is monitored continuously during every session. Predefined stop criteria include physiological instability (such as a drop in oxygen saturation or significant changes in heart rate) or any verbal/non-verbal signs of distress or "cybersickness."

Inclusion Criteria:

• Adult patients, 18 years or older
• Patients in the Intensive Care Unit
• Ventilated patients must be in the post-extubation period
• The patient has a projected remaining hospital stay of at least 4 days (assessed by the clinical team)
• The patient has RASS score between -1 and +1
• The patient has the ability to move both arms, even if with difficulties, as assessed by the clinical team, to be able to interact with the software
• The patient can maintain a stable sitting position (30° to 60°)
• The patient is able to communicate (speech, gesturing, or writing)
• The patient can communicate and understand Portuguese
• The patient or a legal representative provided informed consent

Exclusion Criteria:

• Severe cognitive and neurodegenerative diseases: mental illness requiring institutionalization; acquired or congenital intellectual disability; known severe brain injuries (e.g., stroke with significant residual deficits); moderate to severe Traumatic Brain Injury (TBI) (defined by the duration of loss of consciousness/post-traumatic amnesia or documented residual deficits); diagnosed neurodegenerative diseases (e.g., Parkinson's disease with severe movement impairment, Huntington's disease, severe Alzheimer's disease, or dementia of any etiology that prevents autonomy in daily life at baseline)
• The patient uses neuromuscular blocking agents
• The patient has a positive CAM-ICU result at the time of initial screening
• The patient has active psychotic disorders or suicidal ideation
• The patient has documented epilepsy or history of seizures
• The patient has a "Do Not Resuscitate" (DNR) order, is on life support with exclusive focus on comfort, or has an unexpected survival predicted to be less than 24 hours
• The patient has intoxication by an active substance or withdrawal syndrome requiring ongoing medical management that prevents safe and meaningful participation or accurate cognitive assessment
• Patients with immobility or severe motor limitations in the upper limbs, fine motor skills, or cervical region
• The patient has open wounds on the head or face that may affect the comfortable/safe use of VR glasses, cause discomfort, or present a hygiene risk
• The patient has uncorrected blindness or deafness that prevents the safe/effective use of VR/tablet devices.
• Patients are participating in another rehabilitation study with interventions
• Patients with a scheduled surgery where the ICU stay is expected to be less than 24 hours
• Patients in need of respiratory support

While this is an open-label trial, bias is mitigated through a strict separation of data management duties. The researchers conducting the cognitive assessments do not calculate the final scores or enter them into the primary analysis sheets. Other outcomes are calculated by another researcher or via a stipulated formula. For the full analysis, automated Python scripts are used to ensure objective reporting.