IELTS Academic Reading — Full Practice Test
This is original practice content written for The Band 8+ Blueprint — not taken from a real IELTS test. You have 60 minutes for all 3 passages combined (40 questions total), matching real test timing. Switch between passages using the tabs above — your answers are saved as you go. Submit anytime using the button in the header.
AFor much of the twentieth century, forest ecologists viewed trees primarily as individual competitors, each locked in a private struggle for sunlight, water, and soil nutrients. A tree's success, in this view, was a matter of outcompeting its neighbours. Over the past three decades, however, a very different picture has emerged: forests, far from being collections of isolated rivals, may function more like interconnected communities, with individual trees exchanging resources and even warning signals through vast underground networks.
BThe physical basis for this network lies in mycorrhizal fungi, organisms that form symbiotic relationships with plant roots. In exchange for sugars produced through photosynthesis, these fungi extend thread-like structures called hyphae deep into the soil, dramatically increasing a tree's effective root surface area and its access to water and minerals such as phosphorus and nitrogen. Crucially, the hyphae of a single fungal colony can connect to the roots of multiple trees simultaneously, creating a shared underground infrastructure that some researchers have nicknamed the "wood wide web."
CSome of the most influential evidence for this idea comes from the work of ecologist Suzanne Simard, who used radioactive carbon isotopes to trace the movement of sugars between trees connected by fungal networks. Her experiments, conducted in forests in British Columbia, suggested that older, larger trees - sometimes termed "mother trees" - could transfer carbon to smaller, shaded seedlings nearby, potentially improving the seedlings' chances of survival. Subsequent studies have found similar patterns of resource-sharing in other forest types, though the amounts transferred and their significance for seedling survival remain subjects of active investigation.
DUnderground fungal connections are not the only channel through which trees appear to exchange information. Above ground, many species release volatile organic compounds into the air when attacked by insects or pathogens. Neighbouring trees that detect these airborne chemical signals have, in some experiments, shown increased production of defensive compounds even before an attack reaches them directly. This has led some scientists to describe such signalling, cautiously, as a form of early warning system, though whether this constitutes intentional "communication" in any meaningful sense remains a matter of considerable scientific debate.
ENot all researchers are persuaded by the more expansive claims made about forest networks. Critics point out that many of the foundational studies involved relatively small numbers of trees in controlled or semi-controlled conditions, raising questions about how well the findings generalise to the scale and complexity of real, undisturbed forests. Others caution against anthropomorphising fungal and chemical processes that may simply be by-products of resource competition rather than evidence of cooperation. A tree releasing distress chemicals, they argue, may be responding to its own damage rather than deliberately signalling to others.
FDespite this ongoing debate, the practical implications of network research have begun to influence forestry and conservation practice. Some forest managers now advocate preserving so-called mother trees during logging operations, on the reasoning that their below-ground connections may support the regeneration of surrounding younger trees. Conservation policies in a small number of regions have started to incorporate mycorrhizal network health as one factor in assessing overall forest resilience, particularly in the context of recovery after fire or clear-cutting.
GWhatever the eventual scientific consensus, research into forest networks has already reshaped how many ecologists think about competition and cooperation in the natural world. Rather than viewing forests solely as collections of individual organisms competing for finite resources, researchers increasingly consider the possibility that trees exist within complex, interdependent systems - systems whose full extent and significance science is still only beginning to map.
Do the following statements agree with the information given in the passage? Choose TRUE, FALSE, or NOT GIVEN.
Which paragraph (A-G) contains the following information? Some paragraphs may not be used.
Complete the sentences below. Use NO MORE THAN TWO WORDS from the passage for each answer.
AFor much of the early twentieth century, educators and psychologists in many English-speaking countries regarded childhood bilingualism with suspicion. Children raised speaking two languages were sometimes thought to experience confusion, delayed vocabulary development, and even measurable cognitive disadvantage compared with monolingual peers. Some early studies appeared to support this view, though many were later criticised for failing to account for socioeconomic differences between the groups being compared.
BBeginning in the 1960s, and accelerating significantly from the 1980s onward, this consensus began to shift. Researchers increasingly turned their attention to executive function - the set of mental processes involved in planning, switching attention between tasks, and inhibiting irrelevant information. A growing body of research suggested that bilingual individuals, both children and adults, sometimes performed better on tasks requiring these skills than comparable monolingual individuals.
CPerhaps the most widely cited line of this research comes from psychologist Ellen Bialystok and colleagues, whose studies examined the relationship between bilingualism and the onset of dementia symptoms in older adults. Several of these studies reported that bilingual patients, on average, showed initial symptoms of dementia several years later than monolingual patients with comparable levels of underlying brain pathology - suggesting that a lifetime of managing two languages might contribute to what researchers term "cognitive reserve."
DThe mechanism proposed to explain such effects centres on the idea that bilingual speakers must constantly manage two active language systems, suppressing one while using the other. This ongoing mental exercise, the theory suggests, may strengthen the same executive function networks used more broadly for attention control and cognitive flexibility, producing benefits that extend beyond language use itself.
EThis body of research has not gone unchallenged. Several attempts to replicate specific bilingual advantage findings have failed to reproduce the original results, and some researchers have argued that published studies showing positive effects may be more likely to appear in journals than studies showing no effect - a pattern known as publication bias. Critics have also noted that many early studies did not adequately control for socioeconomic status, immigration history, or education level, all of which can independently influence cognitive test performance.
FDespite this unresolved debate, findings on bilingualism have already influenced educational policy in a number of regions. Some school systems have expanded bilingual and dual-language immersion programmes partly on the strength of claimed cognitive benefits, alongside other social and cultural motivations for multilingual education.
GToday, the field remains genuinely divided. Some researchers maintain that properly controlled studies continue to show modest but real cognitive advantages associated with bilingualism, while others argue that the evidence is, at best, inconsistent, and that early enthusiasm outpaced what the data could reliably support. Both camps generally agree, however, that studying complex cognitive phenomena across diverse populations presents substantial methodological challenges that are unlikely to be fully resolved in the near future.
The passage has 7 paragraphs, A-G. Choose the correct heading for each paragraph from the list below.
Choose the correct letter, A, B, C, or D.
Complete the summary using words from the box below.
AAutomated decision-making systems are now used to help determine outcomes in hiring, lending, insurance pricing, and criminal justice, among many other domains. Proponents argue that such systems, properly designed, can reduce the inconsistency and overt bias that sometimes characterise human decision-making, applying identical criteria to every case regardless of the decision-maker's mood, fatigue, or personal prejudice. Critics, however, contend that algorithmic objectivity is often more apparent than real.
BThe central difficulty is that algorithms are typically trained on historical data - records of past decisions and outcomes. If that historical data reflects existing societal biases, whether in lending practices, policing patterns, or hiring decisions, an algorithm trained on it risks learning to reproduce those same biases, potentially at scale and with a veneer of mathematical neutrality that can make the resulting bias harder to detect and challenge than an individual human decision would be.
CThis concern was thrown into sharp public focus by an investigation into a widely used criminal risk-assessment tool, which was found to produce substantially different error rates for defendants of different racial backgrounds, despite not explicitly using race as an input variable. The controversy that followed illustrated a problem now well documented in the technical literature: removing an explicitly sensitive variable from a model does not guarantee that its effects are removed, since other variables can serve as close proxies for it.
DCompounding the difficulty, researchers have shown mathematically that several intuitively reasonable definitions of algorithmic fairness cannot generally be satisfied simultaneously except under narrow conditions. A model calibrated to produce equal error rates across groups will not, in general, also produce equal false-positive rates across those same groups, if the underlying base rates differ between groups. This is not a matter of insufficient engineering effort; it is a mathematical property of the situation, meaning that designers of such systems face genuine trade-offs rather than a single "correct" technical solution.
EA further complication arises from the internal complexity of many modern predictive models, which can involve millions of parameters interacting in ways that resist straightforward human interpretation. This "black box" quality makes it difficult for affected individuals, or even the organisations deploying such systems, to fully explain why a particular decision was reached, raising concerns about accountability and the ability to meaningfully contest adverse outcomes.
FIn response to such concerns, a number of jurisdictions have begun developing regulatory frameworks specifically addressing automated decision-making, typically requiring some combination of impact assessments, rights to explanation, or human review of consequential automated decisions. The scope, enforceability, and practical effectiveness of these emerging frameworks varies considerably, and comprehensive international coordination on the issue remains limited.
GSome researchers push back against overly pessimistic readings of this evidence, arguing that the appropriate comparison is not between an imperfect algorithm and an imagined perfectly fair human decision-maker, but between an imperfect algorithm and the real, demonstrably biased and inconsistent human decision-making processes it may replace. From this perspective, a carefully audited algorithm, even an imperfect one, may represent a genuine improvement over the status quo in some contexts.
HWhat appears increasingly clear, across both technical and policy discussions, is that fairness in automated decision-making is unlikely to be resolved through technical means alone. Most researchers in the field now argue that meaningful fairness requires ongoing human oversight, transparent auditing processes, and the willingness to revisit and adjust systems as their real-world consequences become better understood, rather than treating fairness as a problem that can be permanently "solved" through a fixed technical configuration at the point of design.
Do the following statements agree with the claims of the writer? Choose YES, NO, or NOT GIVEN.
Which paragraph (A-H) contains the following information?
Which TWO of the following are mentioned in the passage as challenges of algorithmic decision-making? Choose exactly two options.
Complete the summary using words from the box below.
Your Result
Band conversion is an approximate guide based on typical Academic Reading conversion tables. Real IELTS tests calibrate conversion slightly differently test to test.